From: Subject: Stata Press books Date: Fri, 13 Aug 2010 09:30:04 +0200 MIME-Version: 1.0 Content-Type: multipart/related; type="text/html"; boundary="----=_NextPart_000_0000_01CB3ACA.1C64E450" X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2900.5931 This is a multi-part message in MIME format. ------=_NextPart_000_0000_01CB3ACA.1C64E450 Content-Type: text/html; charset="utf-8" Content-Transfer-Encoding: quoted-printable Content-Location: http://www.stata.com/distrib/policies/webforms/statapress.html =EF=BB=BF Stata Press books

Stata Press books

<= IMG=20 border=3D0=20 = src=3D"http://www.stata.com/distrib/policies/webforms/daus2_thumb.jpg"=20 width=3D84 height=3D105> Data Analysis = Using Stata,=20 2nd Edition
Ulrich Kohler and Frauke = Kreueter
Copyright=20 2009
ISBN-10: 1-59718-046-7
ISBN-13:=20 978-1-59718-046-7
Table of=20 contents (from the Stata website)
Preface= =20 (pdf from the Stata website)
Author=20 index (pdf from the Stata website)
Subject=20 index (pdf from the Stata website)
Download = the=20 datasets used in this book (from the Stata Press=20 website)
 
Comment from the = Stata=20 technical group:=20

Updated to include changes to Stata over the past several = years,=20 Data Analysis Using Stata, Second Edition, = comprehensively=20 introduces Stata and will be useful to those who are just = learning=20 statistics and Stata, as well as to users of other = statistical=20 packages who are making the switch to Stata. Throughout the = book,=20 Kohler and Kreuter show examples using data from the German=20 Socioeconomic Panel, a large survey of households containing = demographic, income, employment, and other key information. = The=20 authors describe the Graph Editor and time-of-day variables, = two=20 features added in Stata 10, in this new edition.

Kohler and Kreuter=E2=80=99s book is a valuable = introduction to Stata.=20 The authors take a hands-on approach, leading you step by = step=20 through actual Stata sessions to answer practical questions = commonly=20 asked by social scientists.

They begin with an introduction to the Stata interface = and then=20 proceed with a description of Stata syntax and simple = programming=20 tools like foreach loops. The core of the book = includes=20 chapters on producing tables and graphs, performing linear=20 regression, and using logistic regression. Kohler and = Kreuter use=20 multiple examples to illustrate all key concepts.

The rest of the book includes chapters on reading text = files,=20 writing programs and ado-files, and using Internet = resources, such=20 as the search command and the SSC archive.=20

Data Management = Using Stata:=20 A Practical Handbook
Michael N. Mitchell
Copyright = 2010
ISBN-10: 1-59718-076-9
ISBN-13:=20 978-1-59718-076-4
Table of=20 contents (from the Stata website)
Preface = (pdf from the Stata website)
Subject=20 index (pdf from the Stata website)
Download = the=20 datasets used in this book (from the Stata Press=20 website)
 

Comment from the Stata technical group:

Michael N. Mitchell=E2=80=99s Data Management Using = Stata=20 comprehensively covers data-management tasks, from those a = beginning=20 statistician would need to those hard-to-verbalize tasks = that can=20 confound an experienced user. Mitchell does this all in = simple=20 language with illustrative examples.

The book is modular in structure, with modules based on=20 data-management tasks rather than on clusters of commands. = This=20 format is helpful because it allows readers to find and read = just=20 what they need to solve a problem at hand. To complement = this=20 format, the book is in a style that will teach even sporadic = readers=20 good habits in data management, even if the reader chooses = to read=20 chapters out of order.

Throughout the book, Mitchell subtly emphasizes the = absolute=20 necessity of reproducibility and an audit trail. Instead of=20 stressing programming esoterica, Mitchell reinforces simple = habits=20 and points out the time-savings gained by being careful. = Mitchell=E2=80=99s=20 experience in UCLA=E2=80=99s Academic Technology Services = clearly drives=20 much of his advice.

Mitchell includes advice for those who would like to = learn to=20 write their own data-management Stata commands. Even = experienced=20 users will learn new tricks and new ways to approach = data-management=20 problems.

This is a great book=E2=80=94thoroughly recommended for = anyone interested=20 in data management using Stata.=20

Generalized Linear = Models=20 and Extensions, 2nd Edition
James W. Hardin, Joseph = M.=20 Hilbe
Copyright 2007
ISBN-10: = 1-59718-014-9
ISBN-13:=20 978-1-59718-014-6
Table of=20 contents (from the Stata website)
Preface=20 (pdf from the Stata website)
Author=20 index (pdf from the Stata website)
Subject=20 index (pdf from the Stata website)
Download the=20 datasets used in this book (from the Stata Press=20 website)
 
Comment from the = Stata=20 Technical Group:=20

Generalized linear models (GLMs) extend standard linear=20 (Gaussian) regression techniques to models with a = non-Gaussian, or=20 even discrete, response. GLM theory is predicated on the = exponential=20 family of distributions=E2=80=94a class so rich that it = includes the=20 commonly used logit, probit, and Poisson distributions. = Although one=20 can fit these models in Stata by using specialized commands = (e.g.,=20 logit for logit models), fitting them under the GLM = paradigm=20 with Stata=E2=80=99s glm command offers the advantage = of having many=20 models under the same roof. For example, model diagnostics = may be=20 calculated and interpreted similarly regardless of the = assumed=20 distribution.=20

This text thoroughly covers GLMs, both theoretically and=20 computationally. The theory consists of showing how the = various GLMs=20 are special cases of the exponential family, general = properties of=20 this family of distributions, and the derivation of maximum=20 likelihood estimators and standard errors. The book shows = how=20 iteratively reweighted least squares, another method of = parameter=20 estimation, is a consequence of ML estimation via Fisher = scoring.=20 The authors also discuss different methods of estimating = standard=20 errors, including robust methods, robust methods with = clustering,=20 Newey=E2=80=93West, outer product of the gradient, = bootstrap, and jackknife.=20 The thorough coverage of model diagnostics includes measures = of=20 influence such as Cook=E2=80=99s distance, nine forms of = residuals, the=20 Akaike and Bayesian information criteria, and various=20 R2-type measures of explained variability.=20

After presenting general theory, the text then breaks = down each=20 distribution. Each distribution has its own chapter that = discusses=20 the computational details of applying the general theory to = that=20 particular distribution. Pseudocode plays a valuable role = here,=20 since it lets the authors describe computational algorithms=20 relatively simply. Devoting an entire chapter to each = distribution=20 (or family in GLM terms) also allows including = real-data=20 examples showing how Stata fits such models, as well as = presenting=20 certain diagnostics and analytical strategies that are = unique to=20 that family. The chapters on binary data and on count = (Poisson) data=20 are excellent in this regard. Hardin and Hilbe give ample = attention=20 to the problems of overdispersion and zero inflation in = count-data=20 models.=20

The final part of the text concerns extensions of GLMs, = which=20 come in three forms. First, some chapters cover multinomial=20 responses, both ordered and unordered. Although strictly not = part of=20 GLM, the theory is similar in that one can think of a = multinominal=20 response as an extension of a binary response. The examples=20 presented in these chapters often use the authors=E2=80=99 = own Stata=20 programs, augmenting official Stata=E2=80=99s capabilities. = Second, GLMs may=20 be extended to clustered data through generalized estimating = equations (GEEs), and one chapter covers GEE theory and = examples.=20 Finally, GLMs may be extended by programming one=E2=80=99s = own family and=20 link functions for use with Stata=E2=80=99s official = glm command, and=20 the book covers this process. =

= A Gentle = Introduction to=20 Stata, 3rd Edition
Alan C. Acock
Copyright=20 2010
ISBN-10: 1-59718-075-0
ISBN-13:=20 978-1-59718-075-7
Table of=20 contents (from the Stata website)
Preface=20 (pdf from the Stata website)
Author=20 index (pdf from the Stata website)
Subject=20 index (pdf from the Stata website)
Download the=20 datasets used in this book (from the Stata Press=20 website)
 
Comment from the = Stata=20 technical group:=20

Alan C. Acock=E2=80=99s A Gentle Introduction to = Stata, Third=20 Edition is aimed at new Stata users who want to become=20 proficient in Stata. After reading this introductory text, = new users=20 not only will be able to use Stata well but also will learn = new=20 aspects of Stata easily.

Acock assumes that the user is not familiar with any = statistical=20 software. This assumption of a blank slate is central to the = structure and contents of the book. Acock starts with the = basics;=20 for example, the portion of the book that deals with data = management=20 begins with a careful and detailed example of turning survey = data on=20 paper into a Stata-ready dataset on the computer. When = explaining=20 how to go about basic exploratory statistical procedures, = Acock=20 includes notes that will help the reader develop good work = habits.=20 This mixture of explaining good Stata habits and good = statistical=20 habits continues throughout the book.

Acock is quite careful to teach the reader all aspects of = using=20 Stata. He covers data management, good work habits = (including the=20 use of basic do-files), basic exploratory statistics = (including=20 graphical displays), and analyses using the standard array = of basic=20 statistical tools (correlation, linear and logistic = regression, and=20 parametric and nonparametric tests of location and = dispersion).=20 Acock teaches Stata commands by using the menus and dialog = boxes=20 while still stressing the value of do-files. In this way, he = ensures=20 that all types of users can build good work habits. Each = chapter has=20 exercises that the motivated reader can use to reinforce the = material.

The tone of the book is friendly and conversational = without ever=20 being glib or condescending. Important asides and notes = about=20 terminology are set off in boxes, which makes the text easy = to read=20 without any convoluted twists or forward-referencing. Rather = than=20 splitting topics by their Stata implementation, Acock chose = to=20 arrange the topics as they would appear in a basic = statistics=20 textbook; graphics and postestimation are woven into the = material in=20 a natural fashion. Real datasets, such as the General Social = Surveys=20 from 2002 and 2006, are used throughout the book.

The focus of the book is especially helpful for those in=20 psychology and the social sciences, because the presentation = of=20 basic statistical modeling is supplemented with discussions = of=20 effect sizes and standardized coefficients. Various = selection=20 criteria, such as semipartial correlations, are discussed = for model=20 selection.

The third edition of the book has been updated to reflect = the new=20 features included in Stata 11. An entire chapter is devoted = to the=20 analysis of missing data and the use of multiple-imputation = methods.=20 Factor-variable notation is introduced as an alternative to = the=20 manual creation of interaction terms. The new Variables = Manager and=20 revamped Data Editor are featured in the discussion of data=20 management.

<= IMG=20 border=3D0=20 = src=3D"http://www.stata.com/distrib/policies/webforms/imeus_thumb.gif"=20 width=3D84 height=3D105> An Introduction to = Modern=20 Econometrics Using Stata
Christopher F. = Baum
Copyright=20 2006
ISBN-10: 1-59718-013-0
ISBN-13:=20 978-1-59718-013-9
Table of=20 contents (from the Stata website)
Preface= =20 (pdf from the Stata website)
Author=20 index (pdf from the Stata website)
Subject=20 index (pdf from the Stata website)
Download = the=20 datasets used in this book (from the Stata Press=20 website)
 
Comment from the = Stata=20 technical group:=20

An Introduction to Modern Econometrics Using = Stata, by=20 Christopher F. Baum, successfully bridges the gap between = learning=20 econometrics and learning how to use Stata. The book = presents a=20 contemporary approach to econometrics, emphasizing the role = of=20 method-of-moments estimators, hypothesis testing, and = specification=20 analysis while providing practical examples showing how the = theory=20 is applied to real datasets using Stata.=20

The first three chapters are dedicated to the basic = skills one=20 needs to effectively use Stata: loading data into Stata; = using=20 commands like generate and replace, = egen, and=20 sort to manipulate variables; taking advantage of = loops to=20 automate tasks; and creating new datasets by using = merge and=20 append. Baum succinctly yet thoroughly covers the = elements of=20 Stata that a user must learn to become proficient, providing = many=20 examples along the way.=20

Chapter 4 begins the core econometric material of the = book and=20 covers the multiple linear regression model, including = efficiency of=20 the ordinary least- squares estimator, interpreting the = output from=20 regress, and point and interval prediction. The = chapter=20 covers both linear and nonlinear Wald tests, as well as = constrained=20 least-squares estimation, Lagrange multiplier tests, and = hypothesis=20 testing of nonnested models.=20

Chapters 5 and 6 focus on consequences of failures of the = linear=20 regression model's assumptions. Chapter 5 addresses topics = like=20 omitted-variable bias, misspecification of functional form, = and=20 outlier detection. Chapter 6 is dedicated to = non=E2=80=93independently and=20 identically distributed errors and introduces the = Newey=E2=80=93West and=20 Huber/White covariance matrices, as well as feasible = generalized=20 least-squares estimation in the presence of = heteroskedasticity or=20 serial correlation. Chapter 7 is dedicated to using = indicator=20 variables and interaction effects.=20

Instrumental-variables estimation has been an active area = of=20 research in econometrics, and chapter 8 commendably = addresses issues=20 like weak instruments, underidentification, and generalized=20 method-of-moments estimation. Baum uses his wildly popular=20 ivreg2 command extensively in this chapter.=20

The last two chapters briefly introduce panel-data = analysis and=20 discrete and limited-dependent variables. Two appendices = cover=20 importing data into Stata and Stata programming in more = detail. As=20 in all chapters, Baum presents many Stata examples.=20

An Introduction to Modern Econometrics Using Stata = can=20 serve as a supplementary text in both undergraduate and=20 graduate-level econometrics courses and will help students = quickly=20 become proficient in Stata. The book is also useful to = economists=20 and businesspeople wanting to learn Stata by using examples = that are=20 relevant to them.=20

About the author:

Christopher F. Baum is an economist at Boston College, = where he=20 codirects the undergraduate minor in scientific computation. = He is=20 an associate editor of the Stata Journal and = co-organizer of=20 Stata Users Group meetings in Boston. Baum has coauthored = many Stata=20 routines and maintains the Statistical Software Components = Archive=20 of downloadable Stata components. He has taught econometrics = at the=20 undergraduate and graduate levels, making extensive use of = Stata,=20 for many years.

<= IMG=20 border=3D0=20 = src=3D"http://www.stata.com/distrib/policies/webforms/juul3_thumb.jpg"=20 width=3D84 height=3D105> An Introduction to = Stata for=20 Health Researchers, 3rd Edition
Svend Juul and Morten = Frydenberg
Copyright 2010
ISBN-10: = 1-59718-077-7
ISBN-13:=20 978-1-59718-077-1
Table of=20 contents (from the Stata website)
Preface=20 to the third edition (pdf from the Stata website)
Preface=20 to the second edition (pdf from the Stata website)
Preface=20 to the first edition (pdf from the Stata website)
Author=20 index (pdf from the Stata website)
Subject=20 index (pdf from the Stata website)
Download = the=20 datasets used in this book (from the Stata Press=20 website)
 
Comment from the = Stata=20 technical group:=20

Svend Juul and Morten Frydenberg=E2=80=99s An = Introduction to Stata=20 for Health Researchers, Third Edition is distinguished = in its=20 careful attention to detail. The reader will learn not only = how to=20 use Stata for statistical analysis but also the skills = needed to=20 make the analysis reproducible. The authors use a friendly,=20 down-to-earth tone and include tips gained from a lifetime = of=20 collaboration and consulting.

The book is based on the assumption that the reader has = some=20 basic knowledge of statistics but no knowledge of Stata. The = authors=20 build the reader=E2=80=99s abilities as a builder would = build a house:=20 laying a firm foundation in Stata; framing a general = structure in=20 which good work can be accomplished; adding the details that = are=20 particular to various types of statistical analyses; and = finally,=20 trimming with a thorough treatment of graphics.

Juul and Frydenberg start by teaching the reader how to=20 communicate with Stata, not just through its unified syntax, = but=20 also by demonstrating how Stata thinks about its basic = building=20 blocks. The authors show how Stata views data, thus allowing = the=20 reader to see the variety of possible data structures. They = also=20 show how to manipulate data to create a dataset that is well = documented. When demonstrating analysis techniques, the = authors show=20 how to think of analysis in terms of estimation and = postestimation.=20 They make the book easy to use as a learning tool and easy = to refer=20 back to for useful techniques.

Once they introduce Stata to new users, Juul and = Frydenberg fill=20 in the details for performing analysis in Stata. As would be = expected from a book addressing health researchers, Juul and = Frydenberg mostly demonstrate the statistical techniques = that are=20 common in biostatistics and epidemiology: = case=E2=80=93control, matched=20 case=E2=80=93control, and incidence-rate data analysis, = which can be=20 stratified or not; linear and generalized linear models, = including=20 logistic, Poisson, and binomial regression; survival = analysis with=20 proportional hazards; and classification using receiver = operating=20 characteristic curves. While presenting general estimation=20 techniques, the authors also spend time with interactions = and=20 techniques for checking model assumptions.

While teaching Stata implementation, Juul and Frydenberg=20 reinforce habits that allow reproducible research and = graceful=20 backtracking in case of errors. Early in the book, they = introduce=20 how to use do-files for creating sequences and log files for = tracking work. At the end of the book, they introduce some = useful=20 programming techniques, such as loops and branching, that = simplify=20 repetitive tasks.

An Introduction to = Stata=20 Programming
Christopher F. Baum
Copyright = 2009
ISBN-10:=20 1-59718-045-9
ISBN-13: 978-1-59718-045-0
Table of=20 contents (from the Stata website)
Preface=20 (pdf from the Stata website)
Author=20 index (pdf from the Stata website)
Subject=20 index (pdf from the Stata website)
Download = the=20 datasets used in this book (from the Stata Press=20 website)
 
Comment from the = Stata=20 technical group:=20

Christopher F. Baum=E2=80=99s An Introduction to Stata = Programming=20 is worthwhile for anyone wanting to learn about programming = in=20 Stata. For the beginner, Baum assumes only that the user is = familiar=20 with Stata, and so he builds up accordingly. For the more = advanced=20 Stata programmer, the book introduces Stata=E2=80=99s Mata = programming=20 language and provides optimization tips for day-to-day work. = All=20 readers will find better, new ways to approach old tasks. =

Baum steps the reader through the three levels of Stata=20 programming. First up are do-files. Though often thought of = as=20 simple batch files, do-files support both loops and = conditional=20 execution, and hence can be used for automation as well as=20 reproducibility. While giving examples of do-file = programming, Baum=20 introduces useful but often-overlooked Stata constructions. =

Next come ado-files, which are used to extend Stata by = creating=20 new commands that share the syntax and behavior of official=20 commands. Baum gives an example of how to write a simple = additional=20 command for Stata, complete with documentation and = certification.=20 After writing the simple command, users can then learn how = to write=20 their own custom estimation commands by using both = Stata=E2=80=99s built-in=20 numerical maximum-likelihood estimation routine, ml, = and its=20 built-in nonlinear least-squares routines, nl and=20 nlsur.

Finishing up the book are two chapters on programming in = Mata,=20 which is Stata=E2=80=99s matrix programming language. Mata = programs are=20 integrated into ado-files to build a custom estimation = routine that=20 is optimized for speed and numerical stability. While = stepping=20 through these structures, Baum weaves in the details that = are needed=20 to become an expert at Stata programming, so readers will = also learn=20 more about Stata itself while learning the tools for = programming.=20

Baum approaches each topic by first explaining the = background and=20 need for the topic, then looking at the basic usage and = examples,=20 and finally examining use within larger, more applied = =E2=80=9Ccookbook=E2=80=9D=20 examples. Many of his examples come from questions posed on = the=20 Statalist listserver, so they address complexities of = interest to a=20 broad range of Stata users. The programming examples cover = an array=20 of topics, illustrate some of Stata=E2=80=99s built-in tools = (such as the=20 resampling techniques of bootstrapping and jackknifing), and = offer=20 solutions to tricky data management questions.

The breadth and depth of this book make it a necessity = for anyone=20 interested in programming in Stata. =

<= IMG=20 border=3D0=20 = src=3D"http://www.stata.com/distrib/policies/webforms/saus3_thumb.jpg"=20 width=3D84 height=3D105> An Introduction to = Survival=20 Analysis Using Stata, 3rd Edition
Mario Cleves, = William=20 Gould, Roberto G. Gutierrez, Yulia V. Marchenko
Copyright = 2010
ISBN-10: 1-59718-074-2
ISBN-13:=20 978-1-59718-074-0
Table of=20 contents (from the Stata website)
Preface=20 to the Third Edition (pdf from the Stata website)
Preface=20 to the Second Edition (pdf from the Stata website)
Preface = to the Revised Edition (pdf from the Stata = website)
Preface=20 to the First Edition (pdf from the Stata website)
Chapter 1 = =E2=80=93=20 The problem of survival analysis (pdf from the Stata=20 website)
 
Comment from the = Stata=20 technical group:=20

An Introduction to Survival Analysis Using Stata, = Third=20 Edition is the ideal tutorial for professional data = analysts who=20 want to learn survival analysis for the first time or who = are well=20 versed in survival analysis but are not as dexterous in = using Stata=20 to analyze survival data. This text also serves as a = valuable=20 reference to those readers who already have experience using = Stata=E2=80=99s=20 survival analysis routines.

The third edition has been updated for Stata 11, and it = includes=20 a new chapter on competing-risks analysis. This chapter = describes=20 the problems posed by competing events (events that impede = the=20 failure event of interest), and covers estimation of = cause-specific=20 hazards and cumulative incidence functions. Other = enhancements=20 include the handling of missing values by multiple = imputation in Cox=20 regression, a new-to-Stata-11 system for specifying = categorical=20 (factor) variables and their interactions, three additional=20 diagnostic measures for Cox regression, and a more efficient = syntax=20 for obtaining predictions and diagnostics after Cox = regression.

Survival analysis is a field of its own that requires = specialized=20 data management and analysis procedures. To meet this = requirement,=20 Stata provides the st family of commands for = organizing and=20 summarizing survival data. The authors of this text are also = the=20 authors of Stata=E2=80=99s st commands.

This book provides statistical theory, step-by-step = procedures=20 for analyzing survival data, an in-depth usage guide for = Stata=E2=80=99s=20 most widely used st commands, and a collection of = tips for=20 using Stata to analyze survival data and to present the = results.=20 This book develops from first principles the statistical = concepts=20 unique to survival data and assumes only a knowledge of = basic=20 probability and statistics and a working knowledge of Stata. =

The first three chapters of the text cover basic = theoretical=20 concepts: hazard functions, cumulative hazard functions, and = their=20 interpretations; survivor functions; hazard models; and a = comparison=20 of nonparametric, semiparametric, and parametric = methodologies.=20 Chapter 4 deals with censoring and truncation. The next = three=20 chapters cover the formatting, manipulation, = stsetting, and=20 error checking involved in preparing survival data for = analysis=20 using Stata=E2=80=99s st analysis commands. Chapter 8 = covers=20 nonparametric methods, including the Kaplan=E2=80=93Meier = and Nelson=E2=80=93Aalen=20 estimators and the various nonparametric tests for the = equality of=20 survival experience.

Chapters 9=E2=80=9311 discuss Cox regression and include = various examples=20 of fitting a Cox model, obtaining predictions, interpreting = results,=20 building models, model diagnostics, and regression with = survey data.=20 The next four chapters cover parametric models, which are = fit using=20 Stata=E2=80=99s streg command. These chapters include = detailed=20 derivations of all six parametric models currently supported = in=20 Stata and methods for determining which model is = appropriate, as=20 well as information on stratification, obtaining = predictions, and=20 advanced topics such as frailty models. Chapter 16 is = devoted to=20 power and sample-size calculations for survival studies. The = final=20 chapter covers survival analysis in the presence of = competing risks.=20

Maximum Likelihood = Estimation with Stata, 3rd Edition
William Gould, = Jeffrey=20 Pitblado, William Sribney
Copyright 2006
ISBN-10:=20 1-59718-012-2
ISBN-13: 978-1-59718-012-2
Table of=20 contents (from the Stata website)
Preface=20 (pdf from the Stata website)
Author=20 index (pdf from the Stata website)
Subject=20 index (pdf from the Stata website)
Download = the=20 datasets used in this book (from the Stata Press=20 website)
 
Comment from the = Stata=20 technical group:=20

Maximum Likelihood Estimation with Stata, 3rd = Edition, is=20 written for researchers in all disciplines who need to fit = models=20 using maximum likelihood estimation. This edition offers a = wealth of=20 material about the ml command, updated to include new = features introduced in Stata 9.=20

Noteworthy features in ml include=20

  • constraints() =E2=80=94 linear constraints=20
  • technique() =E2=80=94 four optimization = algorithms=20 (Newton=E2=80=93Raphson, DFP, BFGS, and BHHH)=20
  • vce(oim) =E2=80=94 observed information matrix = variance=20 estimator=20
  • vce(opg) =E2=80=94 outer product of gradients = variance=20 estimator=20
  • vce(robust) =E2=80=94 = Huber/White/sandwich/robust variance=20 estimator=20
  • svy =E2=80=94 complete and automatic support = for survey data=20 analysis

In addition, the authors give advice for developing your = own=20 estimation command and illustrate how to write your = estimation=20 command so that it supports the new svy prefix = introduced in=20 Stata 9.=20

In the final chapter, the authors illustrate the major = steps=20 required to get from log-likelihood function to fully = operational=20 estimation command. This is done using several different = models:=20 logit and probit, linear regression, Weibull regression, the = Cox=20 proportional hazards model, random-effects regression, and = seemingly=20 unrelated regression. =

Meta-Analysis in = Stata: An=20 Updated Collection from the Stata Journal
Jonathan A. = C.=20 Sterne (editor)
Copyright 2009
ISBN-10:=20 1-59718-049-1
ISBN-13: 978-1-59718-049-8
Table of=20 contents (from the Stata website)
Introduction=20 (pdf from the Stata website)
Author=20 index (pdf from the Stata website)
Command=20 index (pdf from the Stata website)
Download = the=20 datasets used in this book (from the Stata Press=20 website)
 
Comment from the = Stata=20 technical group:=20

Stata has some of the best statistical tools available = for doing=20 meta-analysis. The unusual thing about those tools is that = none of=20 them are part of official Stata, so you will not find them = in the=20 Stata documentation. They are all contributed and documented = by=20 researchers in the field who also happen to be proficient = Stata=20 developers.

Meta-analysis allows researchers to combine results of = several=20 studies into a unified analysis that provides an overall = estimate of=20 the effect of interest and to quantify the uncertainty of = that=20 estimate. This collection of articles from the Stata Journal = makes the=20 work of 21 authors available in one collection. Previously, = you had=20 to dig through many Stata Journal articles (and older = Stata Technical=20 Bulletin inserts) to find all the programs. No more! = All the=20 articles are now in one volume, and the associated commands = can be=20 installed at one time.

This is not merely a retrospective collection. Editor = Jonathan=20 Sterne convinced over half the authors to update their = software and=20 articles for the collection, resulting in a much more = cohesive=20 volume. The programs have a more unified syntax than in = their=20 original forms and, among the commands that draw graphs, = almost all=20 now produce modern Stata graphs=E2=80=94they can even be = edited in the Graph=20 Editor.

In his opening comments and the introductions to each = section,=20 Sterne relates how the articles tie together and how they = fit in the=20 overall literature of meta-analysis. He organizes the = collection=20 into four areas: classic meta-analysis, meta-regression, = graphical=20 and analytic tools for detecting bias, and recent advances = such as=20 meta-analysis for dose=E2=80=93response curves, diagnostic = accuracy,=20 multivariate analyses, and studies containing missing = values. The=20 collection addresses both common and complex methods for = conducting=20 a meta-analysis, including implementations of contemporary = advances=20 that will help keep the reader up to date.

The collection includes 16 articles and 15 new Stata = commands for=20 meta-analysis. The articles cover topics ranging from = standard and=20 cumulative meta-analysis and forest plots to = contour-enhanced funnel=20 plots and nonparametric analysis of publication bias. In = their=20 articles, the authors present conceptual overviews of the=20 techniques, thorough explanations, and detailed descriptions = and=20 syntax of new commands. They also provide examples using = real-world=20 data. In short, this collection is a complete introduction = and=20 reference for performing meta-analyses in Stata.=20

Microeconometrics = Using=20 Stata, Revised Edition
A. Colin Cameron and Pravin K. = Trivedi
Copyright 2010
ISBN-10: = 1-59718-073-4
ISBN-13:=20 978-1-59718-073-3
Table of=20 contents (from the Stata website)
Preface=20 to the Revised Edition (pdf from the Stata = website)
Preface=20 to the First Edition (pdf from the Stata website)
Author=20 index (pdf from the Stata website)
Subject=20 index (pdf from the Stata website)
 
Comment from the = Stata=20 technical group:=20

Microeconometrics Using Stata, Revised Edition, by = A.=20 Colin Cameron and Pravin K. Trivedi, is an outstanding = introduction=20 to microeconometrics and how to do microeconometric research = using=20 Stata. Aimed at students and researchers, this book covers = topics=20 left out of microeconometrics textbooks and omitted from = basic=20 introductions to Stata. Cameron and Trivedi provide the most = complete and up-to-date survey of microeconometric methods = available=20 in Stata.

The revised edition has been updated to reflect the new = features=20 available in Stata 11 that are germane to microeconomists. = Instead=20 of using mfx and the user-written margeff = commands,=20 the revised edition uses the new margins command, = emphasizing=20 both marginal effects at the means and average marginal = effects.=20 Factor variables, which allow you to specify indicator = variables and=20 interaction effects, replace the xi command. The new=20 gmm command for generalized method of moments and = nonlinear=20 instrumental-variables estimation is presented, along with = several=20 examples. Finally, the chapter on maximum likelihood = estimation=20 incorporates the enhancements made to ml in Stata 11. =

Early in the book, Cameron and Trivedi introduce = simulation=20 methods and then use them to illustrate features of the = estimators=20 and tests described in the rest of the book. While = simulation=20 methods are important tools for econometricians, they are = not=20 covered in standard textbooks. By introducing simulation = methods,=20 the authors arm students and researchers with techniques = they can=20 use in future work. Cameron and Trivedi address each topic = with an=20 in-depth Stata example, and they reference their 2005 = textbook,=20 Microeconometrics:=20 Methods and Applications, where appropriate.

The authors also show how to use Stata=E2=80=99s = programming features to=20 implement methods for which Stata does not have a specific = command.=20 Although the book is not specifically about Stata = programming, it=20 does show how to solve many programming problems. These = techniques=20 are essential in applied microeconometrics because there = will always=20 be new, specialized methods beyond what has already been=20 incorporated into a software package.

Cameron and Trivedi=E2=80=99s choice of topics perfectly = reflects the=20 current practice of modern microeconometrics. After = introducing the=20 reader to Stata, the authors introduce linear regression,=20 simulation, and generalized least-squares methods. The = section on=20 cross-sectional techniques is thorough, with up-to-date = treatments=20 of instrumental-variables methods for linear models and of=20 quantile-regression methods.

The next section of the book covers estimators for the = parameters=20 of linear panel-data models. The authors=E2=80=99 choice of = topics is=20 unique: after addressing the standard random-effects and=20 fixed-effects methods, the authors also describe mixed = linear=20 models=E2=80=94a method used in many areas outside of = econometrics.

Cameron and Trivedi not only address methods for = nonlinear=20 regression models but also show how to code new nonlinear = estimators=20 in Stata. In addition to detailing nonlinear methods, which = are=20 omitted from most econometrics textbooks, this section shows = researchers and students how to easily implement new = nonlinear=20 estimators.

The authors next describe inference using analytical and=20 bootstrap approximations to the distribution of test = statistics.=20 This section highlights Stata=E2=80=99s power to easily = obtain bootstrap=20 approximations, and it also introduces the basic elements of = statistical inference.

Cameron and Trivedi then include an extensive section = about=20 methods for different nonlinear models. They begin by = detailing=20 methods for binary dependent variables. This section is = followed by=20 sections about multinomial models, tobit and selection = models,=20 count-data models, and nonlinear panel-data models. Two = appendices=20 about Stata programming complete the book.

The unique combination of topics, intuitive introductions = to=20 methods, and detailed illustrations of Stata examples make=20 Microeconometrics Using Stata an invaluable, hands-on = addition to the library of anyone who uses microeconometric = methods.=20

= Multilevel and = Longitudinal=20 Modeling Using Stata, 2nd Edition
Sophia Rabe-Hesketh = and=20 Anders Skrondal
Copyright 2008
ISBN-10:=20 1-59718-040-8
ISBN-13: 978-1-59718-040-5
Table of=20 contents (from the Stata website)
Preface=20 (pdf from the Stata website)
Author=20 index (pdf from the Stata website)
Subject=20 index (pdf from the Stata website)
Download the=20 datasets used in this book (from the Stata Press=20 website)

Comment from the Stata technical group:=20

Multilevel and Longitudinal Modeling Using Stata, = Second=20 Edition, by Sophia Rabe-Hesketh and Anders Skrondal, = looks=20 specifically at Stata=E2=80=99s treatment of generalized = linear mixed=20 models, also known as multilevel or hierarchical models. = These=20 models are =E2=80=9Cmixed=E2=80=9D because they allow fixed = and random effects, and=20 they are =E2=80=9Cgeneralized=E2=80=9D because they are = appropriate for continuous=20 Gaussian responses as well as binary, count, and other types = of=20 limited dependent variables.

The second edition has much to offer for readers of the = first=20 edition, reading more like a sequel than an update. The text = has=20 almost doubled in length from the original, coming in at 562 = pages.=20 This second edition incorporates three new chapters: a = chapter on=20 standard linear regression, a chapter on discrete-time = survival=20 analysis, and a chapter on longitudinal and panel data = containing an=20 expanded discussion of random-coefficient and growth-curve = models.=20 The authors have updated this edition for Stata 10, = expanding on=20 discussions in the original edition and adding new in-text = examples=20 and end-of-chapter exercises. In particular, the authors = have=20 thoroughly covered the new Stata commands xtmelogit = and=20 xtmepoisson.

The first chapter provides a review of the methods of = linear=20 regression. Rabe-Hesketh and Skrondal then begin with the=20 comparatively simple random-intercept linear model without=20 covariates, developing the mixed model from principles and = thereby=20 familiarizing the reader with terminology, summarizing and = relating=20 the widely used estimating strategies, and providing = historical=20 perspective.

Once the authors have established the mixed-model = foundation,=20 they smoothly generalize to random-intercept models with = covariates=20 and then to a discussion of the various estimators (between, = within,=20 and random-effects). The authors then discuss models with = random=20 coefficients, followed by models for growth curves. The = middle=20 chapters of the book apply the concepts for Gaussian models = to=20 models for binary responses (e.g., logit and probit), = ordinal=20 responses (e.g., ordered logit and ordered probit), and = count=20 responses (e.g., Poisson).

The text continues with a discussion of how to use = multilevel=20 methods in discrete-time survival analysis, for example, = using=20 complimentary log-log regression to fit the proportional = hazards=20 model. The authors then consider models with multiple levels = of=20 random variation and models with crossed (nonnested) random = effects.=20 In its examples and end-of-chapter exercises, the book = contains real=20 datasets and data from the medical, social, and behavioral = sciences=20 literature.

The book has several applications of generalized mixed = models=20 performed in Stata. Rabe-Hesketh and Skrondal developed=20 gllamm, a Stata program that can fit many = latent-variable=20 models, of which the generalized linear mixed model is a = special=20 case. As of version 10, Stata contains the xtmixed,=20 xtmelogit, and xtmepoisson commands for = fitting=20 multilevel models, in addition to other xt commands = for=20 fitting standard random-intercept models. The type of models = fit by=20 these commands sometimes overlap; when this happens, the = authors=20 highlight the differences in syntax, data organization, and = output=20 for the two (or more) commands that can be used to fit the = same=20 model. The authors also point out the relative strengths and = weaknesses of each command when used to fit the same model, = based on=20 considerations such as computational speed, accuracy, = available=20 predictions, and available postestimation statistics.

In reference to the first edition, a reviewer for = American=20 Statistician commends Rabe-Hesketh and Skrondal for = promoting=20 the appropriate use of multilevel and longitudinal modeling. = The=20 reviewer writes in the August 2006 issue, =E2=80=9CAll too = often computer=20 manuals leave off ... important aspects of an analysis, but = the=20 authors have been careful to provide a well-rounded and = complete=20 approach to model fitting and interpretation.=E2=80=9D

In summary, this book is the most complete, up-to-date = depiction=20 of Stata's capacity for fitting generalized linear mixed = models. The=20 authors provide an ideal introduction for Stata users = wishing to=20 learn about this powerful data-analysis tool.=20

Regression Models = for=20 Categorical Dependent Variables Using Stata, 2nd = Edition
J.=20 Scott Long and Jeremy Freese
Copyright 2006
ISBN-10:=20 1-59718-011-4
ISBN-13: 978-1-59718-011-5
Table=20 of contents (from the Stata website)
Preface= =20 (pdf from the Stata website)
Author=20 index (pdf from the Stata website)
Subject=20 index (pdf from the Stata website)
Download the=20 datasets used in this book (from the Stata Press=20 website)

Comment from the Stata technical group:=20

Regression Models for Categorical Dependent Variables = Using=20 Stata, 2nd Edition, by J. Scott Long and Jeremy Freese, = shows=20 how to fit and interpret regression models for categorical = data with=20 Stata. Nearly 50% longer than the previous edition, the book = covers=20 new topics for fitting and interpretating models included in = Stata=20 9, such as multinomial probit models, the stereotype = logistic model,=20 and zero-truncated count models. Many of the interpretation=20 techniques have been updated to include interval as well as = point=20 estimates.=20

Although regression models for categorical dependent = variables=20 are common, few texts explain how to interpret such models.=20 Regression Models for Categorical Dependent Variables = Using=20 Stata, 2nd Edition, fills this void, showing how to fit = and=20 interpret regression models for categorical data with Stata. = The=20 authors also provide a suite of commands for hypothesis = testing and=20 model diagnostics to accompany the book.=20

The book begins with an excellent introduction to Stata = and then=20 provides a general treatment of estimation, testing, fit, = and=20 interpretation in this class of models. Binary, ordinal, = nominal,=20 and count outcomes are covered in detail in separate = chapters. The=20 final chapter discusses how to fit and interpret models with = special=20 characteristics, such as ordinal and nominal independent = variables,=20 interaction, and nonlinear terms. One appendix discusses the = syntax=20 of the author-written commands, and a second gives details = of the=20 datasets used by the authors in the book.=20

This book is filled with concrete examples. Because all = the=20 examples, datasets, and author-written commands are = available from=20 the authors=E2=80=99 website, readers can easily replicate = the examples=20 using Stata. This book is ideal for students or applied = researchers=20 who want to know how to fit and interpret models for = categorical=20 data.

Seventy-six Stata = Tips, 2nd=20 Edition
H. Joseph Newton and Nicholas J. Cox=20 (editors)
Copyright 2009
ISBN-10: = 1-59718-071-8
ISBN-13:=20 978-1-59718-071-9
Table of=20 contents (from the Stata = website)
 
About the = book:=20

Since 2003, the Stata=20 Journal has included Stata Tips on special issues in = data=20 analysis with Stata. Seventy-six Stata Tips, 2nd = Edition=20 compiles these useful guides into a compact tome for ease of = reference. In keeping with the Stata spirit, Tips are from = Stata=20 users and StataCorp employees alike and will serve as = guideposts for=20 both new and experienced users. Seventy-six Stata = Tips=20 includes the first 33 tips of the series, previously = published in=20 the book Thirty-three Stata=20 Tips. =

Stata par la = pratique :=20 statistiques, graphiques et =C3=A9l=C3=A9ments de = programmation
Eric=20 Cahuzac et Christophe Bontemps
Droits d=E2=80=99auteur = 2008
ISBN-10:=20 1-59718-042-4
ISBN-13: 978-1-59718-042-9
Table des=20 mati=C3=A8res (from the Stata = website)

Commentaire du Stata groupe technique

Stata par la pratique par Eric Cahuzac et = Christophe=20 Bontemps propose une introduction compl=C3=A8te =C3=A0 = l'usage de Stata en=20 Fran=C3=A7ais. S'appuyant sur des exemples clairs = =C3=A9crits dans un langage=20 simple, cet ouvrage guide l'utilisateur au travers des = diff=C3=A9rentes=20 fonctionnalit=C3=A9s de Stata 10. L'ensemble des outils = n=C3=A9cessaires =C3=A0 un=20 travail sur donn=C3=A9es est abord=C3=A9 : exploration des = donn=C3=A9es,=20 statistiques descriptives, mod=C3=A9lisation, = inf=C3=A9rence, tests,=20 graphiques, ainsi que les sorties pour publication. En = outre,=20 l'ouvrage inclut =C3=A9galement une introduction =C3=A0 la = programmation et=20 propose des extraits de code utiles pour r=C3=A9soudre les = probl=C3=A8mes=20 fr=C3=A9quemment rencontr=C3=A9s par les utilisateurs. Il = contient le mat=C3=A9riel=20 essentiel pour transformer le d=C3=A9butant en expert, la = clart=C3=A9 de=20 l'ouvrage rendant ce processus particuli=C3=A8rement rapide. =

L'ouvrage propose un apprentissage de Stata par des = approches=20 vari=C3=A9es. Pour certains sujets (statistiques = descriptives par=20 exemple), les commandes appropri=C3=A9es sont expos=C3=A9es = et leur=20 utilisation expliqu=C3=A9e de mani=C3=A8re simple. Si = diff=C3=A9rents choix sont=20 possibles, les avantages et inconv=C3=A9nients de chaque = commande sont=20 explicit=C3=A9s afin de guider l'utilisateur dans ses choix. = Pour les=20 sujets plus complexes, l'approche est de type exploratoire : = le=20 lecteur est accompagn=C3=A9 dans l'utilisation des = diff=C3=A9rentes commandes=20 pour analyser ou visualiser les donn=C3=A9es sur la base = d'exemples=20 choisis. Une autre originalit=C3=A9 int=C3=A9ressante de ce = livre r=C3=A9side dans=20 le fait qu'il ne se limite pas aux commandes standard de = Stata, mais=20 pr=C3=A9sente aussi de nombreuses commandes additionnelles = issues de la=20 communaut=C3=A9 des utilisateurs de Stata. Les exemples = propos=C3=A9s sont=20 principalement tir=C3=A9s de l'=C3=A9conomie et des sciences = sociales, mais=20 sont illustratifs pour tout lecteur int=C3=A9ress=C3=A9 par = une application=20 statistique, quelle que soit sa sp=C3=A9cialit=C3=A9. Cet = ouvrage couvre un=20 champ tr=C3=A8s large et s'adresse aux d=C3=A9butants comme = aux utilisateurs=20 confirm=C3=A9s, qu'ils soient =C3=A9tudiants ou chercheurs.=20

<= IMG=20 border=3D0=20 = src=3D"http://www.stata.com/distrib/policies/webforms/vgsg2_thumb.jpg"=20 width=3D84 height=3D105> A Visual Guide to = Stata=20 Graphics, 2nd Edition
Michael N. = Mitchell
Copyright=20 2008
ISBN-10: 1-59718-039-4
ISBN-13:=20 978-1-59718-039-9
Table of=20 contents (from the Stata website)
Preface to=20 the Second Edition
Preface to=20 the First Edition
Chapter 1 = =E2=80=94=20 Introduction
Subject=20 index
 
Comment from the = Stata=20 technical group:=20

Weighing in with 20% more pages than the original, the = second=20 edition of A Visual Guide to Stata Graphics is no = mere=20 update. Author Michael Mitchell adds coverage of almost = every=20 feature added to Stata graphics since the first edition. = Foremost=20 among these additions is the interactive Graph Editor, = introduced in=20 Stata 10, of which the author says

[...] You need to use the Graph Editor for only = a=20 short amount of time to see what a smart and powerful tool = it is.=20 Whereas commands offer the power of repeatability, the = Graph=20 Editor provides a nimble interface that permits you to = tangibly=20 modify graphs like a potter directly handling clay. =

Mitchell adds an extensive chapter about the Editor, = where he=20 first introduces the Graph Editor then shows it in action. = This=20 chapter maintains the overarching style of the book by using = over=20 120 color graphics and screen captures to show exactly how = things=20 are done and exactly how they look on the graph. With = pictures and=20 words, Mitchell shows you how to change the color, size, or=20 placement of any titles, markers, annotations, or other = objects on=20 your graph by using just a few mouse clicks. More subtly, he = shows=20 you how to change such things as the number of ticks and = labels on=20 your axes, the number of columns in your legends, the label = on an=20 individual point, and more. He even shows you how to = convert, for=20 example, a scatterplot to a line plot and how to rotate or = pivot bar=20 charts. Mitchell also covers such advanced topics as how to = draw=20 lines and arrows on graphs so that they continue to = reference your=20 objects of interest even if you resize the graph, combine it = with=20 other graphs, or change the scale or range of the axes. In = short, he=20 exposes all the Graph Editor=E2=80=99s tools, from the = simplest to the most=20 powerful. Mitchell does not stop there; almost every example = in the=20 book now shows you how to accomplish the desired graph or = effect not=20 only by using a command or command-line option but also by = using the=20 Graph Editor. Just look for the symbol to learn how to = produce the=20 displayed result with the Editor.

Beyond the Graph Editor, Mitchell covers major new = features such=20 as time-series axes with intuitive controls for labeling and = adding=20 text and lines; panel-data plots; and local polynomial = smooths and=20 CIs (which join a host of previously discussed smooths and = fits). He=20 also covers more-specific new features such as options for=20 controlling aspect ratios and for changing all text sizes=20 simultaneously.

The book retains its visual style, presenting the reader = with a=20 color-coded, visual table of contents that runs along the = right edge=20 of every page and shows readers exactly where they are in = the book.=20 You can see the color-coded chapter tabs without opening the = book,=20 providing quick visual access to each chapter.

The heart of each chapter is a series of entries that are = typically formatted three to a page. Each entry shows a = graph=20 command (with the emphasized portion of the command = highlighted in=20 red), the resulting graph, a description of what is being = done, the=20 dataset and scheme used, and, new in the second edition, a = section=20 showing how to produce the result by using the Graph Editor. = Because=20 every feature, option, and edit is demonstrated with a graph = or=20 screen capture, you can often flip through a section of the = book to=20 find exactly the effect you are seeking.

Aside from inserting a new second chapter about the Graph = Editor,=20 Mitchell retains the original organization of the book. The = first=20 chapter discusses how to use the book, the types of Stata = graphs,=20 how to use schemes to control the overall appearance of = graphs, and=20 how to use options to make specific modifications. He also = outlines=20 a process for building graphs using the graph = command. The=20 second chapter is a complete overview of the Graph Editor. =

Mitchell advisedly spends the most time in his next = chapter,=20 which is about twoway graphs such as scatterplots, line = plots, area=20 plots, bar plots, range plots, regression fits, and smooths. = Mitchell shows how to create each of these types of graphs = and how=20 to use options (and the Graph Editor) to control how the = graph=20 looks. He also introduces graphing across groups of data; = options=20 for adding titles, notes, etc.; and options for adding and=20 controlling legends. Beyond the basics, he shows how to = easily=20 overlay plots to obtain such graphs as regression fits with = error=20 contours and observed data scatters, local polynomial = smooths with=20 scatters of their underlying data, stock-market-style graphs = of open=20 and close values with quantities traded as a bar chart at = the=20 bottom, histograms with density smooths, and the like. = Because=20 Stata=E2=80=99s graph command will let you customize = any aspect of=20 the graph, Mitchell spends ample time showing you the most = valuable=20 options for obtaining the look you want. After reading this = chapter,=20 you will have a thorough grasp on how to create graphs in = Stata. Or,=20 if you are in a hurry to discover one special option, you = can skim=20 the chapter until you see the effect you want, then glance = at the=20 command to see what is highlighted in red.

In the succeeding five chapters, Mitchell covers = scatterplot=20 matrices, bar graphs, box plots, dot plots, and pie charts. = As with=20 twoway graphs, he shows you how to create each of these = graphs and=20 how to adjust every aspect of the graph to your taste (or to = a=20 publisher=E2=80=99s required form).

In chapters 9 and 10, Mitchell undertakes an in-depth=20 presentation of the options that are available across almost = all=20 graph types=E2=80=94options that add and change the look of = titles, notes,=20 and such; control the number of ticks on axes; control the = content=20 and appearance of the numbers and labels on axes; control = legends;=20 add and change the look of annotations; graph over = subgroups; change=20 the look of markers and their labels; apply schemes to = control the=20 look of the graph; change the look of graph regions; size = graphs and=20 their elements; and more. Again, he now shows how to make = these=20 changes both with options and in the Graph Editor.

To complete the graphical journey, Mitchell discusses and = demonstrates the 11 styles that unite and control the = appearance of=20 the myriad number of graph objects. These styles are angles, = colors,=20 clock positions, compass directions, connecting points, line = patterns, line widths, margins, marker sizes, orientations, = marker=20 symbols, and text sizes.

That completes the main body of the Visual Guide, = but=20 don't skip the appendix. There, Mitchell first gives a quick = overview of the dozens of statistical graph commands that = are not=20 strictly the subject of the book. Even so, these commands = use the=20 graph command as an engine to draw their graphs, and=20 therefore almost all that Mitchell has discussed applies to = them. To=20 make this clear, he shows explicitly how to apply common = options and=20 common Graph Editor tools to statistical graphs. Second, he=20 addresses combining graphs=E2=80=94showing you how to create = complex and=20 multipart images from previously created graphs. Third, in a = crucial=20 section entitled "Putting it all together", Mitchell shows = us how to=20 do just that. We learn more about overlaying twoway plots, = and we=20 learn how to combine data management and graphics to create = such=20 plots as bar charts of rates with capped confidence = intervals,=20 scatterplots with range-finder confidence intervals in both=20 dimensions, and population pyramids. Fourth, Mitchell warns = us about=20 mistakes that can be made when typing graph commands = and how=20 to correct them. Fifth, he show us how to create our own = scheme=20 files. Scheme files allow you to control every aspect of how = your=20 graphs look without having to specify options. They are the = answer=20 to department or journal standards or if you just want all = your=20 graphs to have a common appearance that is not one of the = schemes=20 shipped with Stata. As with the rest of the book, this = section=20 includes cross-references to the Stata Graphics Reference = Manual to provide more depth on the subject. Finally, = Mitchell=20 reviews all the datasets, schemes, and other online = supplements=20 available for the book.

The second edition of A Visual Guide to Stata = Graphics is=20 a complete guide to Stata=E2=80=99s graph command and = the associated=20 Graph Editor. Whether you want to tame the Stata = graph=20 command, quickly find out how to produce a graphical effect, = master=20 the Stata Graph Editor, or learn approaches that can be used = to=20 construct custom graphs, this is the book to read.=20

<= IMG=20 border=3D0=20 = src=3D"http://www.stata.com/distrib/policies/webforms/wdaus_thumb.jpg"=20 width=3D84 height=3D105> The Workflow of Data = Analysis=20 Using Stata
J. Scott Long
Copyright 2009
ISBN-10:=20 1-59718-047-5
ISBN-13: 978-1-59718-047-4
Table = of=20 contents (from the Stata website)
Preface= =20 (pdf from the Stata website)
Author=20 index (pdf from the Stata website)
Subject=20 index (pdf from the Stata website)
 
Comment=20 from the Stata technical group:=20

The Workflow of Data Analysis Using Stata, by J. Scott = Long, is=20 an essential productivity tool for data analysts. Aimed at anyone = who=20 analyzes data, this book presents an effective strategy for = designing and=20 doing data-analytic projects.

In this book, Long presents lessons gained from his experience = with=20 numerous academic publications, as a coauthor of the immensely = popular=20 Regression=20 Models for Categorical Dependent Variables Using Stata, = and as a=20 coauthor of the SPOST routines, which are downloaded over 20,000 = times a=20 year.

A workflow of data analysis is a process for managing all = aspects of=20 data analysis. Planning, documenting, and organizing your work; = cleaning=20 the data; creating, renaming, and verifying variables; performing = and=20 presenting statistical analyses; producing replicable results; and = archiving what you have done are all integral parts of your = workflow.

Long shows how to design and implement efficient workflows for = both=20 one-person projects and team projects. Long guides you toward = streamlining=20 your workflow, because a good workflow is essential for = replicating your=20 work, and replication is essential for good science.

An efficient workflow reduces the time you spend doing data = management=20 and lets you produce datasets that are easier to analyze. When you = methodically clean your data and carefully choose names and = effective=20 labels for your variables, the time you spend doing statistical = and=20 graphical analyses will be more productive and more enjoyable. =

After introducing workflows and explaining how a better = workflow can=20 make it easier to work with data, Long describes planning, = organizing, and=20 documenting your work. He then introduces how to write and debug = Stata=20 do-files and how to use local and global macros. Long presents = conventions=20 that greatly simplify data analysis=E2=80=94conventions for = naming, labeling,=20 documenting, and verifying variables. He also covers cleaning, = analyzing,=20 and protecting your data.

While describing effective workflows, Long also introduces the = concepts=20 of basic data management using Stata and writing Stata do-files. = Using=20 real-world examples, Stata commands, and Stata scripts, Long = illustrates=20 effective techniques for managing your data and analyses. If you = analyze=20 data, this book is recommended for you.=20

= ------=_NextPart_000_0000_01CB3ACA.1C64E450 Content-Type: image/jpeg Content-Transfer-Encoding: base64 Content-Location: http://www.stata.com/distrib/policies/webforms/daus2_thumb.jpg /9j/4AAQSkZJRgABAgAAZABkAAD/7AARRHVja3kAAQAEAAAAPAAA/+4ADkFkb2JlAGTAAAAAAf/b AIQABgQEBAUEBgUFBgkGBQYJCwgGBggLDAoKCwoKDBAMDAwMDAwQDA4PEA8ODBMTFBQTExwbGxsc Hx8fHx8fHx8fHwEHBwcNDA0YEBAYGhURFRofHx8fHx8fHx8fHx8fHx8fHx8fHx8fHx8fHx8fHx8f Hx8fHx8fHx8fHx8fHx8fHx8f/8AAEQgAaQBUAwERAAIRAQMRAf/EAK0AAAEFAQEAAAAAAAAAAAAA AAUAAwQGBwIBAQABBQEAAAAAAAAAAAAAAAAAAQIDBAUGEAABAgQDAwYGDgkFAAAAAAABAgMAEQQF IRIGMRMHQVFhIhQIcYEysnM1kaFC0iNDY7N0FRYXJxixwdFiosIzUzSCwyREVBEAAgECAggEBQIG AwAAAAAAAAECEQMSBCExQXHBMlIFUYJEFWGBoSITQxTwkdHhIyRCYoP/2gAMAwEAAhEDEQA/ALR3 nOImtNJ3WyNaduDtG3VMOqqEtpCgpSVgAmYMOihknQg8F+Jmub7YbjU3i5O1D7NYGmlrSAQjdJVI YDlMYfec1csuOB0qma/asvC6pYlWhof2pvmUf8pXtRie6X+o1vb7PScnVV+GPaldOAg9zv8AUL7f Z6Rfaq/Sn2pUpdEL7nf6hPb7PSI6rvmWfalDxD9kHud/qD2+z0nKtWX2WFWr2oPc7/UC7fZ6Rs6u v4H+WrpwEKu5X+oX2+z0nCtYahCZirVKWGAhfcr/AFB7fZ6TlestQBP+YoHwD9kL7jf6g9vs9IT+ 29x+w3bs57bkyb7lz7M0dLjl+3xV+7Br+Rz+Bfnw00Yisd4J55u52kNrUibLk8pI91FfuMmmqGr2 WKcZVW0BcP3HXLRVFaiuVTtUZ/Fpjns828NfjwNfClLQthZ8JDwxninnh58IBKnivJkOTxQ9CVG1 rkPEZwqQoyt3yYWgo2XZjHn2wqQDa3OrIfshaANuu9U+DGFSBE2f4bz/AHo6/wBN5OBynqfPxOe8 N6ztHoXPOiDuWtGp2TllvA3Ddsrs1ZL/ANP+2mOezv8Ax+fA1bkqS+RbNwRtGyKAzGE7Rp5dehTr jm6ZScoIEyTtwh8Y1KmYzn43RKrIt3tDtDVIZJ3u8luVAYmZlKXPOHYR9jMK5Guqg5XaQulPQrqS UKKElS2Uk5kgYnkkZRM7EkqkdvuEJSw6SrurlKctkRmiiOXZbdhhRxwp4ePZC0EGnHeqrZswhyQB ef4ZTjrfTeTgcn6nz8T3vDes7R6Fzzor9y1o1Oycst5E4TM7yy1+GyqHzQjn87qXz4F3NypNbi6u UmE5SihQrq4S7ZcVUKFNKb3jSiVAAyIPjh8J0IL9nG61oyDda56rq2akJybmRaRtlIzx8cOxaakt m0oRcfEl3PWIct7jTVOpNQ6koUokZEzwJHPFmeYxKlCC1kKTTb0IFaPstruK6lysQHls5QhgkjAz moylOK5Zz2YnCijoqV7WVFR229OU1GfgciVluc8ilbUz9uFjpLeSuSnbrLWAN7gJRJQtHK3pjDph yQhYp/hbPonHV+n8nA5L1Pn4nfeG9Z2j0LnnRW7lrRqdk5ZbyXwIoUVNluhUZZKtGEp7WhFGGTV7 S3qF7tdcZrdxNBudE2wnDEkTnKUoqZ7KxtrQULF1yALmBnLwxkF9EZ4EpmNkocSRBlWnqbNpnDkT RYFqXn6de9YcU05KWdCikyPSIeidJNUeklaQ07SX+qq1V7yzuQkltCuusqn1iTPAShLs3FKhDm8w 7SWFayv6mt7dpvNVQsu75pkjIoyzAKE8qpcoia1LFFMs5e45wUmC1PdU+zE1CUtk/wAJ5/ufqjqP T+Tgcj6nz8R7vDes7R6Fzzoq9y1o1Oycst4U7vvqO7/S0fNCG5Lle8j7zzx3cQzxdq71TWalXaV1 Tbqnyl5dE1vnQjculAy5HOqXg2CZRFnIxclipSm3f/QzbVdhnNTqbiKmpBXbnW2KVwB0IonFl1Rp 3QUnrSUkPNpUFJOMx4Iy1lrFNel/H4rgXfyzqd3LUesjQadqGbe8HjmqL4wy1JOQOJZDZC+uJ5y5 JPWw5obCzZrNNr/rp+f9iVzn9rpvAtw1br9dGGjbFN1DiF5nWaV0ZczIWkoJUsZm3DlIIx9qJYZa xWuLRvXj/Qd+W7TVp3fAGV+qdVPtrDVKlaErqKbe07S1kuMCWdKplCUlXj6DD45a0npfg9L8SVX7 uxeK0fAkWK93wioeez0dQ24UNuNpW1NEuQqM1dOERX7MFSmlULtluaamtvgP1rNc0EuVTTre+mtD jiVDPPGYJ2xHFrYWYyT1ERb3VPgh6Qpdp/hDP5OftR036Hl4HJep8/Ek94b1naPQuedFTuWtGn2T llvCnd99R3f6Wj5oQmS5XvI+888d3EsPFC7XS02qmrbfMKLjjTig0XpTpX1M9UA+VUIbSPDLliS5 lY3ZfdqSMlXHFaClVfEXVGdLabOd+065Sugh7dE/BJStag3MFOdRKUkS2EbDFXNZC1GNdRJauybo V1jX12uVLdkppm6CspaNNQxnDrikLWrL8IjJMj3QKQZco2xkzycIOOnEnKmz6GtC9KSexpAhvX1/ TlQ9TmoDu5SzvGlNrIUpSXF5mxu1CYBBG0Yyxwllkbex017f4Y6GYn4V1DVo1RcK59NJV06UKRT7 xx9CFNjegiacqgB7rk5oS/loxWKL2lnLXpSeFrYPu1BQ4FynlIUJ4jAziNRL1Avq3XRv1FT03ZAw WlBa15s01ASknDBMRWMtgdakFjLfjbdSpLdwI6It0LJoM/wbn8j/ACx0X6HlOR9T5+JM7w3rO0eh c86KnctaNPsnLLeFO776ju/0tHzQhuT5XvI+888d3E0W8LvyENGzIpnHs/w6atS0J3cj5JQFHNml FtNVMZlb1BVa+TbnCti2JPxYS6+QcFTzTSOiKHcsDgq1pUsZVPFoM4o1XBuofdq2KRlx45lPUs94 4oE4uFQB2dMc/cwtJJy+Z0lqMk22l8hqsrlHYo8vLCRgWUgHXVmblMWIQHDFrpmrjc2aN14MNuqI Lp5MJyHJM8kPuSwRbSEk6KpH1BQs265u0bL/AGhtuRDmE8RORlhMQ6zNzjVqgsHVVBK3DyRMkONJ n+C8/kJ/wx0H6PlOQ9T5+JP7w3rO0ehc86KfctaNPsnLLeE+7+ZWO7/S0fNCI8o6Re8Z3nnju4mo lcl+KJnPSZFCv60qg1ac0/dH9EZ+flWKLmSj95jFTcjNUjgPBGVGB06QMfrSpPLM4xNGAoPdeLiw lPlKIAnhiYkSoKdXu01dpqUU9SUqWtAcBQZiRwhLN1TVUJF1BpQsgnExNVD6HKmjl2HbAmLQ0mR+ 5aXLuP5Y6Cv+HynHeq8/En94b1naPQuedFPuWtGn2TllvJ3AZeWxXb6Wj5oRTtzpEO7Ks47uJpRe +Fl+7+uEd7SZijoKpxJqt1Y0Hncl/CYrXpYi7kY/eY0/brklsLdaUgLM0TljyxFE3Vdi9QJfeUlR SfKGEhzxKkSIt9u4Y6hfpQ7UUrqHlSKG5oIkRhjmhuLwKM89BPQ9BXrxbXKSr3S1KW8CUqzEmRBl LxQVLlmeJVI/ZinqqEuWGuRYiqni2gROWAhFIlwF9y/g5l+Rl7UdNX/B5eBxFP8Ab/8ATiTO8N6z tHoXPOip3LWjR7Jyy3kngYZWO6fS0fNCMuboh/c+dbuJohc+HKZ9bJOXROIHIz6aCNX0lNVuUyKl CXW0qUoNrAUCch5DzQ1uo6LarQD6rtNCjT9UaWjbFSciGd2gBeZSwAEy55wqJbNx41V6DObbw5u3 aBUVdO6iZzKStuSUDlJUTyQ9zNGecjSiZtdMkBtoDYEpE/AIWCMSRWaPh9bah1de+7ndfKlFKkJU EzUcEzMTQsYtpZlnZRWFbCuaj0Rva80FACG21AuvlBPIDLqg44xWu26SojQymewxxS/kVu7aOeow G8/wq8GGspSpZ6M0sOmIXVazUsZ1T07NvwDHZnfur7N8bl3f+rZHVV/1/JwORxr93XZj4gHvYalu VnvFhRRpZIdp3SvfNhwzCwMJnCLM8vC5zKpVtZu5a5HSpjdm438Q7Ky4xa6qmpWnl7xxKaVszUBl n1p8kRPt9l7PqxLmduzdZOpN/MTxX3u++smN5lyZuys+TOctnPDfbbHT9WR/uZ6qiX3ieK61trVc mCtokoPZWcCoSPJzQe22On6sP3M/EN0PF/ixdrYmqf1FbqZrfBIZfpm0rm2C4lwAJxTmSB4TB7bY 6fqxVmZraT3eKHFlxxumXq60KRUtOKWezoASElCSk9TaredXoSYPbrHT9WCzE1tO6rinxfomahSN WWp9NK0FtBilQvekJJ3aOp5QygY88Ku32fD6sR35+JVmO8lxdZbS21c6dKE7B2Rk7TPlESRyVtal 9RrvSesK6Z498Ybrc1UTV+oqJTwW8p56jZylaQBLAbSBCxysI6q6fiDuyesZ1Zxp4p2287mrutuu VSllIL4oGFBKSScgC0mR5TEc8hZm6yVWSRzVyKwxdEaf9eXD7iPr2bf1h2Ttk92nd73Jn/pSy5Z+ 52RN+OOHDT7dQ3G64q6Ql3j+FOo9aVtprbU/TttUDLjbrb28zqUtQUCnIlQlIcsPToMaqYn+XTXn 9ym9h33kLiEwi/Lprz+5Tew77yDEGEX5dNef3Kb2HfeQYgwhuh4UcYqGkapKa5MIpmBJlooUtKRh MDM2dsoMQuEkOcOONrlWurXdac1LrYacc3RmUhe8HxUvKM5wVDCdN8OuN7a0rRdmEqR5Emzh1iqf 9PnMFQwlaPd114STvKbHod95BiEwno7u2vUmaXqcHnG+H6EwuIMIld3bXyjNT1OTznfH9KYMQYTb fsFfPuX+y8k9v7J2XPJW7z5MubZmy+KGjqaDaq3yIQUhQAKABQAKABQAKABQAKABQASP+vAB/9k= ------=_NextPart_000_0000_01CB3ACA.1C64E450 Content-Type: image/jpeg Content-Transfer-Encoding: base64 Content-Location: http://www.stata.com/distrib/policies/webforms/dmus_thumb.jpg /9j/4AAQSkZJRgABAgAAZABkAAD/7AARRHVja3kAAQAEAAAAPAAA/+4ADkFkb2JlAGTAAAAAAf/b AIQABgQEBAUEBgUFBgkGBQYJCwgGBggLDAoKCwoKDBAMDAwMDAwQDA4PEA8ODBMTFBQTExwbGxsc Hx8fHx8fHx8fHwEHBwcNDA0YEBAYGhURFRofHx8fHx8fHx8fHx8fHx8fHx8fHx8fHx8fHx8fHx8f Hx8fHx8fHx8fHx8fHx8fHx8f/8AAEQgAaQBUAwERAAIRAQMRAf/EAJ8AAAEEAwEAAAAAAAAAAAAA AAYAAwQFAQIHCAEAAwEBAQAAAAAAAAAAAAAAAAMEAgEFEAABAwIDBAMKCwQLAAAAAAABAgMEABEh EgUxEwYHQSIUUWFxkbEygtI0CIGhwUJSclMVFhcYYiMzRtHxosJjJESUVSYnEQACAQIFBAEFAQEA AAAAAAAAAQIRAyExQRIEUYEyE2HwcZGxIhTh/9oADAMBAAIRAxEAPwCR7znNXmDwnx5F0/h3VHoc JyA26tptIUkuFxwE4g9AFajQy6h97vPGfFHEvLlOp69MclzzOkNb5wBKt2gIKRYAbMxqe7Jp4DIr A6X2yR9M0v2SNURDb1CcX3EF5VkqIA71ehbinFMglOSk1UntvylD+IaGkbUmOBcn6ZrlEaqxxPaP pmuUR3E2s/8ATNcwO4mqu0DYs12iOYjD7stDLi0rN0pUR4QL12iOVZTfiuR+GvvG3762X0tl/HSh w/xAlBn9ZCVHKMSkHyin2siS+8RQAkRUhICRmVgAANveqTleXYo43iSKlKCG4Mk2/Q4kH4Rga9Hj SrCnQhvqk69SwZlRUWS46hBOwKUAa7KaWbNwhJ5IlNS4jgO7dSvKAVFOOBwpXui8mO9UloSW3Gle YoHwUKaZzY0bqUhPnG1dckCiaKW2R5wo3oNrI0xTfZnhcXDaj8RrW9HHFgF/InpfLWTQUa/7d6Iq i1kR8jMUH2VPhV5aj5fn2KeN4j9SlBWcRQnJWluhkkPt9dogkG6cbYUyGNY9f2Yng1Lp+gIgcUdl jBM51idqCDYtFOVYQcUgDHrXpSeGJfC25PCqQX6Jq2qPoSuPHs2oABbriADtvc2ubVlOaf8AK/Jy 5CCzf4QTJ3trrSUqPTYgbNuFUY6kuBsu5QVIcRcY2uR5a08sDiGpDjYGdCPBfHDpsKxKmZqKYPan PbaS9ks0VJICkKKspUk2K/2ScMemkxzwwHbXTqU11/l5mw3lr26M39dX6EeoV6/7d6Iqm1kRcjMU H2VPhV5aj5fn2KeN4j9SlAq6Bz3ivTFaXKW+1G38eSrMgBIJQ5j3firtxV/r8/f/AKasSo9reWX2 JXDursoDbDjrMVTYyFtxhaUDObgXuU3I8dZK5Rbxo33DyHKbWOuTvML3WMpvjgL9+mRl1zJJR6Dq HEqw3aFuDG234b0J1ONEN+VkSVLWAUglYQsIsn6SlG1Y3fSN0AHijWITmnSiXVtsuNlD7xXnJUTb K2gWK7H51cRUouJJ6n5YXucm58751sm3w2q3Q83UMdf9u9EVTayIuRmKCLxkgbcx8tR8vy7FPG8R qBrGkai2+5p86PLbjLLclbDqHA2tOKkryk5SLdNTNFBKacbdaQ80tLjLiQttxBCkqSoXSpJGBBGI oAamw48yMuPITmbWMR3O+K7GVDMo1Ob63oeo6AtxTMM6hEWP4i3FkC6thQL7AeihwoqrFfr7lNm/ ilJ7X9ZFro3G+lyzeXKajvoGMfKpLmGBTnIxOFhjWHSWY92ZLxVQm+/UIS0yyHFOPgKbRkGYJ865 UnvUbmsEJ9Vat0wAviXiTWN86w0W2g8AFMPqBJJOGW524EG9ZSdXXUstwtpJvToVGm6FrevRy9nM QNlSVLLSDZDYsQCoXJwOIwFNhCuLwX1kScjkJPbHFhbux+Xm66LZL97ZVehCFev+3eiKptZEfIzM 6cSGGyBchRNvhqPl+XYp43icrh8sdPnvMMxuKY0lxD3aNzGaQVKjRnloI6rqlZkuPLacXsxy2pfs +Bu0J+X/AC6k8JSJTq9QTLbkRo8ZDLTBjoQIwISrLnWCopNqzOdTqjQNMiu4aWaMKazpKVIzJO0E V1NrFHGk1RlI/oWiObxxlCLtuFDwbyqCXAcQRjZQvj00+2lN0axFynK2qpkb8LxQpLSJb6EFKlrI V1lAkAJJ25Ug9UdFM/yqtKm/90qVoh2DwloaDm7OFhBCU5gDspdyEbbolV/JmN6d3FvAuHW224jq G0hKQhVgPq0mUm3ibjFLIFv5C+H5ar0FhTr/ALd6Iqm1kR8jMzp5swggXIUTb4aj5fl2KeL4nIkc l9OW1OfgcUM5FIUZLgaStDLbrvaVG6XBku4lTgN9vgrHs+Bu0sHeSmoZQ5G19YOZUkgNqSpbykKT nS7vFZM4XZRscK57fgNpVaHykRLSI87iOKdQQdxK0uItTqWVIdW9uEpDo6uTrFBTgRfZXXc+AUSx j8kEKecH4gVIZaKGnY+VZDeVYdWyQHbBLmbMpKhe9jXPb8BtDHg3g9PC+jv6WmSmWqTJ7Sp4IyKK lNNoWV9ZWYlTRN+/TbKcpbtEJvNKNOoQ9lVv0n/DI/tCqa/0JS/kwwMhU2rA3um/Tep+VB13DuPJ JUMyfZnvqK8hqMpBT+Qvh+WrdBQU6/7d6Iqm1kR8jMzp5ysIV3FE+I1Hy/LsU8bxOZHknN3mZvXE oILiEFLCwQwWltoFg7lLn705l2+rbG6/b8DdpiPyg1vTdb0eZE1ISo8eYzJnJdLjeXdOJcW42nMr M44lGT6uHgPYmg2j+p8nZUmZNlp1WO1vpMiRHZVFUW2hKQ6l023lwte962WyMM2XMo0KelAaNOHe UmoafrenT2dXCmIT6JLtmFofkLEZuOvM5vD1VBu+I6casjapjIlldrgjrMSCEAdXGtSkEIEvs6c4 P7JHx0uuI2mBEmQkrBwpkZCpwKeWp6Mw8DdbWRWB2jA7KxcsKWKwZyF5wzxQPXH4Bv0Xv8dYoU1C nX/bvRFUWsiPkZigG0dB7iifjqPl+XYp43iC7XLbTUJYvqM5bsZUlbDxdspKpbwfd83LcZrgA9BI pG8dtIrPKxLbzL54j1V2QwkoQ644FEpJSrKrujq7K77PgNo0/wAt48llmINa1EuRXHXO27wb1S3g lKs3QQkI6vcq2xDDcyO7cq6LIdb5MMuzmp6uJtYD7KVpaVvgSkOCywlRFwFf01uTNQQWcMcCjQtR XP8AvnUNQccZLCm5joWjLnzpVa3nJxAPcpbY0KcL1w6aLSCK6jJU6tHBiv4fMV5DTIsTOIF2/wCg W79K1KNAq1/270RT7WRJyMxQfZU+FXlqPl+fYp43iP1KUDMp0tskp89XVR4TTrNvdKgq9PbGooMf KAK9GTIoRLhrKgDMQPDSZMpih4udW6ReluQxRGjmzkhKsDfbgaXqMHN6bXIt3Aa2pGHEg6g+yYrw UbdRXg801tTRiUGA2H4C71/lrh0Kdf8AbvRFUWsiTkZig+yp8KvLUfL8+xTxvEfqUoBzijWzAkxW kqQnNdS1ObANgt36dG44KqzYv1K5Kj0Ms65LLrQRJYDd7Sd4sIKb+aEgdJxrrvSeo+HGjR4OuhdN Sw6cVMvMBYzKzYpuersJxrDlXo0GynVMnsOspwb3R6oC+ve19taTWlDLT1qSEOjKpQDYXhayjsFa T+xlohzVoAUQlpV7C6nCBc49Hhpc+35Nw7lVNmFOnS8hYSUoJSM5y47cyjs71cjLDQ1sq1mD+b/z 7Nhsvtw8dVaE+pzH3l+b3HPB3HUbTdBlMsxHYLb60Ox23TnUtaScywTsSKbBugmcU3icoR7z3OFC cqdSjBI6Oxx+n0a5O2pOrOxe1URt+qHnH/ycb/ZR/VrPoia9jJWk85OZvE70p6ZrekxXITQcSZ0Z hAc61glsZSCq9clai6HYzarTUvY3H/NFaiocUcOpF1ZitiLvLi/zSi5vbDvGj1R6G/dLqxnT+cHO CTIhxRxJocbti5KFqdYioQyqGSBvjk6u8y3bPTR649DPsl1J8LmzzbfXla4y0FrMSFqcjR0JuFBN r5MSb3HdFHqj0D2PqUWue8Tzr0h9LK9ZgSELCsjjESKtJCT3k4YG9dVtHHNkrhDnlzg4onPw/v8A 06EppveJVIgsEOG4TlTZG21cdqPQ6rj6lbxRz85saPqczRzq8CYhqyXHWYLAQoqTc2BR0XtXVajT IPbKuZ6L7S7+UHa8N92LfXsMufc5/N2Wv0VygVAz3j+SfFPGfF0fXNMkRkRWobcbcub3eZ0LWonq pUm3XFsa0pUMtVOQ/pk49+3i+J71K7vDaL9MnHv28XxPepRvDaL9MnHn28XxPepRvDaL9MfHn20X xO+pRuDaL9MfHn20XxO+pRvDaL9MfHn20XxO+pRuDaIe7Hx4Nj8XxPepRvDaZT7s/H6fNkx0+DfD yIo3htEr3ZeP13vIjEnC53x/uUbg2nqD8N6l+WH3Pl/zXZOz9Ns263d+7a9YNHQpvRQBEoAVACoA VACoAVACoAVACoAl/wCnoA//2Q== ------=_NextPart_000_0000_01CB3ACA.1C64E450 Content-Type: image/gif Content-Transfer-Encoding: base64 Content-Location: http://www.stata.com/distrib/policies/webforms/glmext2_thumb.gif R0lGODlhVABpANUAANfo9VBrjufn5yUyTzVMbbjN4pexyKK5z8LU5JKbqXWQqrTI3Iuku212hwUH Dk1Xacrb6neTwB0qRmVrdOf0/sXFxvz8/BEaM7O0tjA0OXWAlMfa6ZycnsLN3dbW1xciOjlBT6zC 1/b29qSmqlhhcaWrtEdOWYWKlrq9w7/W6oyRnEVgg8XX5oSfzN/t+YCEjFt3m6ywucje7+77/8/h 77zQ4cnZ6GWBoT1WeJa83Co/XsDR4MnS43mUwRIXHf///yH5BAAAAAAALAAAAABUAGkAAAb/wJ9w SCwaj8ikcslsOp+WqHRKrVqv1yMGxO2CJpmweEwuPyozF20NWbMh8LgtTq/b7zSRlDhyOHw+fz4g foWGh4iFJB0UNCyPjxAbk48IlpeYCCwbLJmelywCAnp8gKaAJomqqz4nAC6dkDabs52ekHCQurux Ox4epEMjp6apgavIhhkxMzQ7vCy1mrGbk9bX2LUsHRgeo0QcxKgOFwQSyegOEzSw07GzmpaQ2PT0 jx0lFd9DHIiDDjBkLEiw4lw6VRJGUIDwDBq1avUi3huhTwQRDCYyajShIcCCOQAA1FBA4NjBQxNe YdolD2JEe9so7hPCo4VNAzYZFCiwYIEm/xqvAKQ6iSgDIxY7QFWa9lJizIpEOkSYGqFH1RAHDoQo sGmNCxhEE/lQsVBTLlvzmlLSVUCmxSFSq/aw2uMAAwMHfLYBauBD2EQkQiLINa2lS5i6OriNOnUu Xbt3F9TYC3RBA4MgGBD4u0jktBqFQWm6xgu0YqhwG8+tyuBGDwMhJrsB6qLADRIFGiXQcdDHAAkE YqSoUaDGAdDFY0sugIDTJl4hEJye+SOu4wgtNOiAwSD6bAg1WKQICWANgBQa/Io95mPFBx8XFADo WfzSTp7M5SWtxDz29LdCWLdaCw1cgIMCByAw204LhEBDcRsssAENNRjwgCl/mJDBBx9kQP9Ih37o oEEKCPBEH34LcCUZfQvs0FOJBfzHmFxWEejAANwtgBQCNSwHgQwpbFCiJQukYMAFpnzARQaBBALC hr45EBh4OxFn2CU9pljDDj32uFhqNGLXgAMSBNBdT2jWMIcMbHCCwA4bMOCHSYcMMgAhIAQSWA0h oJnUJ/HIE4KLXwaoWo0FDmBmVlnB1mckknSyQ1J9zUlnIQOEAd8xD9Ag0JaAfqJiodUdKqYD291l gAEM3JXXaG5OWsOsBmzGCiCImNDclpNO+maoRJIqYI03XCDBCgq02moPd/X5zCVcEsdcATAAQkK1 B4FQwwazcukrsCwsIKypLSggwQc6rHD/gwI3BIAsA3k9O+u8DBbXanE9GISOtizMCyqwCIiLmqFh ttADAQN8IIEOOAxwwQDIalVclieamCJPWxnAWzomkDjvn7dcIiOY1zG7AsIfIDmnovA2GAJWL7OI Zk8vFxCCreiQIAOP/94CycgEOzaXATecrIN6gXyALF4HrMp0zDO7DDUJhVx6iAIu7FCfO7xsAHSp Yb7GrtHvkQOcAk6zqurLfUZNc2wMXIpr1X8c8GCLXD9Si9fjhm0XuwHgoIMEFxj7rrLMuto2mhb3 lAJYgiTzgd317bKBDZOA13fJTTOggALuInzBBwcy0MPn67aa130Tg8ZzCg1gaPUftGvr/wys9EDw 9bBXdX5D0ScnPAABMIBOQLrJrj4rzTb3aUkBG6ty6QMAQPBIRLpvvloEPR0QAAFFwxC8wgMMDt+x CPbJ0wHtMiC+ATTYkEJBrMxZiHzWY7+7qRH0p8MNMCgaDhB2tLmRA2I30AkDDsQoHAQAMgZYAQ4a 8IHv2U96DADASxiiPav0jwUFeGAAjIcDCTrMJKNLVwRCcDIEMUoSMAgADAywABkoIAUKGAA6JKCg DcTBh3HYX5j6ZwMGBAAB67qBDgIXAB38pmwXSFcCiRaAxNFwAwfAQfGAcgD53EBwOZOBDO7AwYGB 7ToRAI2ZIMC+AJjpdyUUHOF8U0VWsf/LNasKAQsMkABnFGAHzKKBawqQAXTg7w476OBUuKKJSb2s c55r1wCPFkVkgU4HDEPbAVDUIPqYBQAHuIDk7IZIRRJRUrJaQNNYdTrxCe43CzPHHz5AALSxbZUS C88kAJAAdDxgjHZIpBl5R0RLyIo4DVpVq9glvuMRrmp0hJfTWoWXrTRnAzLQADp8wAAaBNOUBYBH r6L1SLW1MnCiO8bD6risZLnKJ48IgXqSMYACeJMOwqQOMQsgJE80qEchUKbnFOBKJ4rSB+hbpviK 1ywGHUBfyWjAPeOQTwCdcXsFsB4kNGGDA3SARzZb5TLbZTTCVfJ3ASDcAHBwg2pihX3/B7kAKSlq ysn8yDkb2MEBqtGJkCrzdCR1JocGkLCVrQBejbrLAw6ygjH+rINWqYEzfDJGGjTodm9CSjLNSdDQ DU6UVTubsgbaAxBIoJDI4CaFMAZV7tEgBHfh53wWYNXYbKA8jvCp2pgpwa8W4gM4YhZBdUCAALTU MskgAHEeqchyaYABLxhBAhKwgBJooAHJQsECYtCnGJSAAZ/93Oc0EEMJOvOgElBUAwIwAEAsLIEM mKcqNJC2xhYotSAgam5TNjrAZiBTRP0tcH9L3ExJ4Ljn4i2HCsdc8mVAAptCEnPhUzgSKKuxGjAW UY/7RA6lLLVEDa94x0ve1CaXuYUj/194edvc0THXBInDrsLOldv29ha849VBbslbPvFiMrnsVe9v 0EvgwoFgBSOU77lIgAIUgMBYoyOcwgSckAlMALwgcKIEVKCCC3D4Aw24MCwTNuH5srfA38XkDRR8 gRI0ICMJwAAJSoCCCZwAAxOIwQs0gIIGxEADl/WACkaA4xh4IAEXkOwFNPACGr/gBD3WAAYakIAR nCu9BS6cb5bI4hJMYAT54HAMfIyCIaNAAiTwrGQTMIAKNKACCSjBCE6A5ASowAFVxoAGKgBnFaQ5 BhUwAWDnm2UkARYHLE4AB2LQ4AY0YMpOfrQEUODZGHCgyiN4cwOgHIM6d2DGRH5zBf9OkOMSANqs vxkwiv9wXA4Mk1zpOZcOVPCCL9SaBBkhwQlAQAIQa+ABJAC2jR/QgBcM4MkPuMADaB1sXJOABCow QQM0QIIJwNK8J+Zt0j7gan3C2lgmJYd0s3xi9Co3ZRNOL4fOu9x1I/e4yx0dUdVDy25blHfZMZZ3 A+ze3rr73QCfL3f5u12AGxze+D3GCrDbWwK3O71XFnjBEX6u1M53vQfPOMDn7QcQMDxlE0iABDgw Aeo+7AXM1YAJ5H3tfQ9ABb9JwJjX3YCjldi7OnjBvnducT+M/NUFK1DKOCCAB4ggAQ9IAK8xgAIT qOABP9YAzCVAZWInoAEPKIEHUuv/gReHWAMCOMGmH0DnB0idBC9IuoVV8Gu2w9sP3Ab6dQjkXhVg gAN3jwGUUfACDDBdsiPowAl+0wAU0DgBKMDACSrwGw9UANoieECZK0BpFBC5AiqYAKX3jniKXKbi f7D3jOZeoMJxQAVCjkEMFoECBqPgBGnecQVyW4Lax0DzNWb8ADxg7ROIAASKr4CNS9DkwqugwZoX Po4TwHjAmiPu3i5YdlMWYj8XW8Z9F/sIJgBkDozgNyVIwAn8zHYMlCC1McAAh1XQAA5PdtPtHzL8 UcBkaGtAshUnrA5ET7IBTZ/chWNSEvYwFTdx3HWA2wVYyJUwC3ht18ZdEscw/Bc0//4ncAfnbhh4 gBlngA9oXhuocU60f3Lnf/hFcOK1cSMWXiWogiY4Xu+mgu+GSSIYfXOnASa4XzuXgzq4gzzIg+rG WwuDSRN4UXRhAApgZ92BAiFQASHQAQlAAJOEScNzPISFAwNEAFB4hTK4hVzYhVx4PClIVJhEAEPI Oz3wUnkBABTgAmw4AwegKCdzPCsgQSXkLnMoQ+5ihVi4h3y4h3p4hX24h05UPmBIWGVoKme4Kllx V+ThAh0gAYKDSSUEhQg2hwiWh1DIh1uYhVbYiXOoh/oXglJYPlJIhiNIFwHFNCzgAuTRCMcliXTo Rncoi1koiISFhQjmhbioLt8DiP/9JYYyeIhhg4YGUANsyIYUkADowjCBc4mziE6AeDy2GAN+FmwP cI0TQAJSRwAVIHUJoAE/NgErsFrAaIo0uBpn+FI09AouQAHQMzxLVIfu4kYxNEDRaIvH1QAjYHkc sGkyR2kaAAI/hgGTdXgqEH6jaI73hoiukhUO0o4AUBC4aIV4KEP1qId+CIZPJAE2Rmoh5mjTlnkD AHtip2sqIHaOJoPlI4xz1y6M8jLJAgPoQgCfaJEBJEm1mImZhAOANTjlBnHuhiQ6sBtBGF5SOIML WTAMoAPdoRUMYEIrFUc2+TswYI90KDi1FB0LQECDiFwKODyypgNIwgAWwAIFsYX/pIiUo3cdZrkC bFOV5sAwErQCMBBDSXQyn2iFS+QTa2gDCxAAPJdqGnYuTxQfCMBLxtKHhniKVkEDBjAA8KIAfbUw ljiVJOVGdUgAGuACs8CGDFEA7uGBDThw4YUDCzADJAAfy0gAD8CaLHkdEAAA27EbJ7MwUlmXN1mV smhaDEABMxACNnCMakgDOGAs/7VuOKeSHwADMzAmfhBFMqiQa7kaLDAD2nEAhFU+cUSXdZlEVmiJ A2QAM0ABoHQAaniMM7ABBEBLlORvgCVeKWMAzgl3W/iaq4EAFBACEsAAiZKXlkmHlUgABrCGrFge x4ieC6Bh6LaDqdVbhxCEatl//9RJnoVVLIX1nQHUnfN4hwTQm414oCBKAQzQXQDmXQsoYdBlCKRo n1aBAO2oHTzZiQ6EmwDkRvNYS+QZFCC6o18xYSVqouQzVLLlG9IpoS3ajjdDmd85lTaKmTCgo2wI FCHBhuyIjBtwNPDmcAVYcRBFDhFKgfdJARSwA8UZRyW0UDFko5bookEREmxiHkDxIDLQmcoIb+Xm oyRWNobwpUTYojp1A8dlWjS5UM3oVQcAC3hFHngVPxBQADkAGwaQAyGApeR2onpqCD93jnSxAEVj DpmYhQiWpnYIPqyYAz3ypmIEADJQAOXhqDkAJFuhABEWb0DJXWBVJyzKLO4yOP9GQ5OzeIl6aACO yqo0YCSrwk850AI5wBOSmgI2wxME0HDj5p7P5A/2iR1P2TDpgoXbWVp2+D/LCiRUCiQFIAN4NQmu KgOsCCQKYGLZllznIhbXajI4AByx+Ksl9InLmgJrOKXIyIpU6gKrWkPtaCQ5MDg/+m+pZTU+cK3Z IUeeaIfiI4vfkwD9Kpw8eowPYk85qobFYqf+9oK3WjVDWBM2cbItcAAEVJndCUCiGgA0MJ5iOrMz ULM2e7M4244z8AMhcIINOIhOpDAqAwhDiAHXeLTYaAJcsBEa8QBMawIWFrVSO7VUW7W45gVYm7VL 8iQmMIT9ID1NQjtiSzuz8xf/dSIO63EK0GdR4SAObvu2arshcqspHMIkhtYhYVC37wG3PqAkXuC2 awsOfDu4GJIBE/ACE6B+mRcOD5Bp6vACL6CPHEBqX/YCguC2DmACc8ZhHFBIxBC4/EC4otsAHNCP IwC50JYBJsB7PjABKuB9I8ABiGtjd8a3GeBoFtZr4gC6QtC2osu3T9IhHfIkeYK3gDC3HsIkmmK7 IGACz5Ynu+u1vzu4mSt22Rhis+sDxZYKDxC5YLB9LyC6e7u30WtGvju9gEtkr3sCsgu5GJABrjYB DoB6JcABD+ABpYu+cMu7P3C++nsKhgtszuu0w/sBzku82PgAEwBs/wu40tvAlmhLtpcrthhyQRBc vtTxtRe8wRxMDA7AvxjxJGUwwiRcwiZ8wihMwh43MBYgAgJQAZZWujI8wzRcwzZ8wzicwzpMw0Tm DQDSwgLgeH43xERcxEZ8xEicxEq8xEaMAhXgDRYwBFHgwr9QxVZ8xVicxVq8xVzcxVosClFABC0s AmQsCmZ8xmicxmq8xmzcxm6MxmSsB1EcBAA7 ------=_NextPart_000_0000_01CB3ACA.1C64E450 Content-Type: image/jpeg Content-Transfer-Encoding: base64 Content-Location: http://www.stata.com/distrib/policies/webforms/acock3_thumb.jpg /9j/4AAQSkZJRgABAgAAZABkAAD/7AARRHVja3kAAQAEAAAAPAAA/+ED4mh0dHA6Ly9ucy5hZG9i ZS5jb20veGFwLzEuMC8APD94cGFja2V0IGJlZ2luPSLvu78iIGlkPSJXNU0wTXBDZWhpSHpyZVN6 TlRjemtjOWQiPz4gPHg6eG1wbWV0YSB4bWxuczp4PSJhZG9iZTpuczptZXRhLyIgeDp4bXB0az0i QWRvYmUgWE1QIENvcmUgNS4wLWMwNjAgNjEuMTM0Nzc3LCAyMDEwLzAyLzEyLTE3OjMyOjAwICAg ICAgICAiPiA8cmRmOlJERiB4bWxuczpyZGY9Imh0dHA6Ly93d3cudzMub3JnLzE5OTkvMDIvMjIt cmRmLXN5bnRheC1ucyMiPiA8cmRmOkRlc2NyaXB0aW9uIHJkZjphYm91dD0iIiB4bWxuczp4bXBS aWdodHM9Imh0dHA6Ly9ucy5hZG9iZS5jb20veGFwLzEuMC9yaWdodHMvIiB4bWxuczp4bXBNTT0i aHR0cDovL25zLmFkb2JlLmNvbS94YXAvMS4wL21tLyIgeG1sbnM6c3RSZWY9Imh0dHA6Ly9ucy5h ZG9iZS5jb20veGFwLzEuMC9zVHlwZS9SZXNvdXJjZVJlZiMiIHhtbG5zOnhtcD0iaHR0cDovL25z LmFkb2JlLmNvbS94YXAvMS4wLyIgeG1wUmlnaHRzOk1hcmtlZD0iRmFsc2UiIHhtcE1NOk9yaWdp bmFsRG9jdW1lbnRJRD0iYWRvYmU6ZG9jaWQ6cGhvdG9zaG9wOjVmY2RmYThmLTMwYjYtMTFkOS05 OTI1LWRiMjI3N2YxOGRkZSIgeG1wTU06RG9jdW1lbnRJRD0ieG1wLmRpZDoyQUZDNkIxNTdDODcx MURGQkQ2OUU2MDlGNEQ3RkY3MyIgeG1wTU06SW5zdGFuY2VJRD0ieG1wLmlpZDoyQUZDNkIxNDdD ODcxMURGQkQ2OUU2MDlGNEQ3RkY3MyIgeG1wOkNyZWF0b3JUb29sPSJBZG9iZSBQaG90b3Nob3Ag Q1M1IE1hY2ludG9zaCI+IDx4bXBNTTpEZXJpdmVkRnJvbSBzdFJlZjppbnN0YW5jZUlEPSJ4bXAu aWlkOjAzODAxMTc0MDcyMDY4MTFCMzRCOEYwMzBDQkZEMENGIiBzdFJlZjpkb2N1bWVudElEPSJh ZG9iZTpkb2NpZDpwaG90b3Nob3A6NWZjZGZhOGYtMzBiNi0xMWQ5LTk5MjUtZGIyMjc3ZjE4ZGRl Ii8+IDwvcmRmOkRlc2NyaXB0aW9uPiA8L3JkZjpSREY+IDwveDp4bXBtZXRhPiA8P3hwYWNrZXQg ZW5kPSJyIj8+/+4ADkFkb2JlAGTAAAAAAf/bAIQABgQEBAUEBgUFBgkGBQYJCwgGBggLDAoKCwoK DBAMDAwMDAwQDA4PEA8ODBMTFBQTExwbGxscHx8fHx8fHx8fHwEHBwcNDA0YEBAYGhURFRofHx8f Hx8fHx8fHx8fHx8fHx8fHx8fHx8fHx8fHx8fHx8fHx8fHx8fHx8fHx8fHx8f/8AAEQgAaQBUAwER AAIRAQMRAf/EAKsAAAEEAwEAAAAAAAAAAAAAAAQAAwUGAQIHCAEAAwEBAQEAAAAAAAAAAAAAAAID AQQFBhAAAQIDBAUGBwoLCQAAAAAAAQIDABEEIRIFBjFBIhMHUWGBkbEUcTLSczUIGKHRQoIjM1MV FhfB4VKSwmPDJCVGJ3KisoM0RISURxEAAgECBQMCBAcBAAAAAAAAAAECEQMhMUESBFFhBSITcTJS FPCBkaHBQmIV/9oADAMBAAIRAxEAPwB/1nOKfELKmfKTD8uYo/R0TlA26tptIUkuFxwFVoOoCGVB XUt3AfPeb8wZBTiON1zlVXGsqG964AFXEBF0SAGiceH5TkXLc0oOiodViCaxOiHGcSA+eVHiz8jy PqOlWY9ASqzDiqEkpfUIj/0uT9bKx48OhA1ebsyImUVaxyCyKryPI+pnZa4Vp5ohKvPWcEzKK9xO oWDr0R1WfIXtZHq2fGcd5xBE8RM5bozrXVOkyASkWS02ckd650qZnS/Ecavyqgm+IGd3FqvVzqLs tAEpdUcl3yNxZMSfi+LFYRQSM95wIl39wctg96OZeSv1+Y45cDjr+oNV53z8m+tFe6EpSo3ZDUOW UdlvnXdWPDhcWmMUdK+11X9iPrS3vXd/G1393p/Oj6Kvpr2Pj6Lf23fyAZ+P8ZTo+aTqHKYcIZG+ XkNqw1JUBO+u2wa48fyNtSmq9C0HQlu7siwtkk6BMR5320HoUU2Mu0lKfGZs1zlKGXFh0KqbB3MM w82FpBPJZOKrjw6FY3ZIFXhGEqmdw2eW0Q6sw6HRHkT6jJwPC0oKhSoIHJKGdqNMin3U+plvBMP1 0Ikbb2zI9URdlP8AqLLky+obewbDmk3+6yKbR4tlsSdiKxoSd+T1HX8PZLLyghCpoVJMiNXLOOqN tE/eYNL+nt2WqUo9vT8jyNQzP3ppPmk9phxoZG+XlAYYJrCTvFWEA6+ePN5i9WdMCsUyZTSPVK1K Q8htLexMpmSZBRlKUhbEIcaqru/YbfTQz9UPlCgaxJJ0zQTr8MUXG/0arvYycFqt7e70icpfN/jh 1xe4yv8AYwcBqJn95bBP6r8cb9t3GXJ7fuL6kqEoI7wgzEhJuUuicH2/c18jsbIwmqQ2E94bugam yP0oPY7iO9XQFqsGfdQoGpSARqQffic+OmniYruOQ2vA30Urv7ykgNq+AeTwxi466h73YhJH7A3d c5T6Y9TQ49QrP3ppPmk9phh4ZGcvGnGHp3qSo31Gzwx53M27sVoXhu0JLFVYivLGMfUrRcxZxDrd Am+G5PONpQld8+LcKr3RDWJQcUTmpJlLrMK4rs4NlhDTbtXimGtvM4gU1YQ27UNvtpYqahd5svNr p0rJBSbTakG0dNIk/UBvtcf6uiCXl7lXyRUKZNI08QlxK17V5VpAKdMiI30hSRbuG6OIiG6tjOAT u2EU7eHq+TUtRSiTilOIUpSybLxUPGnKMlTQaNS5KCCDYJaDMQjHqxslFwASPJYdXhhHQMQXZuqk NmRsnq8ERrgNqavuk0jo3ezu1ax+TyRN3cMgpiVP+Q/jfhj09DmCc/emk+aT2mHHhkOZccbThibx ltr7RHHyGtxRJsseGPNll4ggjen/AApjbcltFlF1Ct83MTNk7YZtAos5BW4bxwvVjVDVOihRWOPU qEO0pdU33zfJAcd2pFvZuKN27sxROIrjIzi1Px/fpF0aH9426wymofZVRsuhxTSSvcqTdu3XisOn 8m7u7ZwViG1m1Wnj3W09XSuXqZK2H2qZ2mVRIXvg2AhalklSUqUDu7ts/GkIysQoxlun9YJphVNS qDaJOqQpw0jqwCpS2tpZWd4VEhyeyE3buuMrEMTrTaHNwhTgAeLaS6B+VdF73Y55IdM1qQoUrpFu wqfUYjJNYm1Kf/Ifxvwx6ehAJz96aT5pPaYYeGQJhtWGMNQN3fN5Zn0xzX41ZaC7k7gNWXKJxRRK byrOhMZCNEZNY5kpNREwIGgQklxOgShHKg4it4W6bZQjmbRENm6mxyuyvitJg7zjGKPMKTROtLDT gcBBF1arEz0TjFco8TJxwwKPX4Xxdr8qPs1a3HMWRiSHqVtqoapyqnS0QUOKYLM2S6RIBYWNO1K6 W92NexJp0AlYRxvQ4aYVL5oWnS8F95pit4ivS8lKXCgLSncXkEKslZzQvuwMozomUFZrbyooZr2s YC6i9ItE7kqJaBLMkEhJlYIac0sjEmRX8ifG/DHboSCc/emk+aT2mGKQyA8Pl9XI/tL7YnMdLEca rqqkC007pQhRvKSUhQnonb4Ii5ULRt1CkY9WBKJ1UlS2vkk+9HPO89C8eKI5gqp/6qf+UnyY5Z8i fRFFxDdONVqxeTVC7Mj5tGr4sc0+XcWiB8ZIaVjdcNFQOb5NPvRCXOudF+PzB8ddyPxXNeNUTAeY QqtUFALZaS2F3T8IXhIy5IRc243R7V+pOVlLqQmI8S8xMNoUjCqkpJ1pYBsnqnze7DW+RNv5ofox PbXRktSZ/r63Cl1ClLpVFK0ll9tAXYDbszEjGy8nKDo0n8AXHqEXz93m8nbK9Pn0x9Ru9Fex59MQ 3P3ppPmk9pig8MgbClgULaCmc1LkemEkx0gpxqz5qYjmmXh8TW8UgJ7sOS0fijlnGp0r4jykNynu 0/miOScRVN9RhSigWNiXIJCOaUCil3Bn3TuzsyA02iIO0bUrtbihbWoahaJyi0OLUSTIHFccLjSJ WyUbOiUUhwqMTcJnFgcMIlvFhCylIMjPmnZELnC9Q0ZVL1M/dlPXup9N2Pqqeinb+Dy9SRz96aT5 pPaYoNDIGwm4aFtJO1eWQOmJXCsUG99QzsuBVmuUcrbLK23kYermCkETIInGbahsYGvEkJJF1XVy xOVsZQYO5izFoUlVtlkiZ9cQlaRVW5Az9cCyVhJCTaCRptlohIWcRZJp0KVjNUSslvZ0gabY7rdp E5Mr9S4ZWqvWzEvdjo9onuG+9ltsqQZoUFTGgicI7Ju47BP+l0/1E/7kdOhx6knn700nzSe0ww8M gbCQo0Td0WzXI9MSuLEomg/vKz8k+0LNColTRjYZpkfWN1bar7KQpB0iM9umQ25PMb782DccAQsm 2fJFVAnU1eXNJ2EyJOkjmjJWqjKXcjKyqKmCDIFuwp1EcsLG0a5FPxcgiaQQFTsi8YCtlffSoXbu 0JTJ556DD0MI59TiEOlCZhaTORlqOsxjiajt9v3Vc/dv2cYc+pK5+9NJ80ntMaPDIbwcfwpCpfDW AekQrNeYUtSHUXHBI6lQtAAXnn6VW0kqbGgwLA2lQd1FPVJCrslaR4TFIupjVCNrV1FO2fhImbJW ykIYEQrtbfQsAELnol0GMGIutTfmqRKRZPk5jGgRD6U6gALTPm5o1ARtS0i64qWhKjdAtJlKUawR 2X/yz/j/ALOJETmXrO8UM25Vz7S0GDusIpnKFt5QdYQ6q+VrB2lapJhkhdzRyun9ZTivTsbhqspE tTKrvc2TadOkQ20xzEfWV4rnTW0n/TZ96DaG42HrJcWn1IY77Rm+QhIVSMATUZWkiyCgbiyV3ELi xSd7U5mPAld1QXEFAZUHkhBX8iQNrRd8NkYbuYzScTOK1bUVrJx3BmkURQC480ynehxClXmk6VJS UyUeeANzI/MnEHiFhFIqpXjGD1snUslqmZbUs3ws3wJeILkidSjKBG72Vn7689SUN5SSUJKHdkRt DPcZL5Wz3nbMdU/TCuw2jLLe8Cn6ZICpmV1MtcDNU2AZk4i5wwnFKjDzVUFUppICn2acBJJTMi06 U6IA3s9Ub5z7nd/ZvO47zRZe3F7R4YmBTfWP4KZqznm6nxzDKimRStUbdNunN7vL6FrUTspUmW2N cMnQxo5D7Mufvp6Xqe8iN3GbRezLn76el6nvIg3BtF7MufvpqXqe8iDcG0XsyZ9+mpep7yINwUF7 MmffpqXqe8iDcFBezJn0aHqXqe8iDcG0Xsy5++npep7yINwbTZPq0cQU+LUUyZ6Zb4diINwbTCvV m4gLJvVFMSbJnfH9CM3BtPUH2bxL7sPqe7+9907vrle3W7nyynCjHQ67QIAAoAFAAoAFAAoAFAAo AFABmAAz/bQAf//Z ------=_NextPart_000_0000_01CB3ACA.1C64E450 Content-Type: image/gif Content-Transfer-Encoding: base64 Content-Location: http://www.stata.com/distrib/policies/webforms/imeus_thumb.gif R0lGODlhVABpAOYAAEtIVTcnZ4qCtKOaw3txqKmkyJOLulVIhouDqWVZkmxsb3Zspri4uDQmV/n5 +Topcerq6pCFrHl1l4J5rtbW1p6XwUMze5iNtoV9sNLU5KKXvnRqm7u50yEhIWllgpSJssjI3Gtg mp2dnUE0bEIzc4aEi1dUaIB4mks7gXFriZmRvQcHB15RiDk1RZuTtEk7eZaPu11YcpeMteXo8mRe eoV7opWKszY0OpyRuZuQuLWzz1JFelJOYTwuZpGGr5KHsKalup+VvoqLsPLz9ZmOt4h+qo2FtpGJ t392rHVwjp6SvE1IYUA9S0U3dCooL0AveIuKkJqTvkg7ccG/18XFxU5Ae6+ry4h+o0Y3ej42W5ya vpORtvb2+CIgKn9/o1BBhJ6UuJaLsjEuOKenp3BloJWSqIiAsj8wcGFUjjAjVI6PtiwmRaCct6Og xO7w9N3e6CYfPIF3ppKMqjQsShoZG5SNuJaWu1pNi4+IuI+Ht0VCVb2w4GBgbkY2fz0scgAAACH5 BAAAAAAALAAAAABUAGkAAAf/gA6Cg4SFhoeIiYqLjIh8N5CQJVZtBQVWVjo6U1Mcnp6amAWVHGwK JydCVhxTIKwgGW8zbkNcXENDEG8gDFZlJSAgoZhWowMVQVEqMMwqKhVWQ4NOf9XVfCpmeQYwFQMD opaWbcdRMHUqOh56HjRJWwU6nxkgsrWCtm5vGUBAJTRAprSpQNBcnSMCzGCYgKQhkgkTMESB4MbB kBvWrhlAgsGMkSPMBgwkSNLgETwimZho5yXOFk2wMsSixQXfkBn0dACB4qGEwChAl51MiGEhxKIY BOCgAMGBGzEZ//ARQKBhRDMCyOU5eeRkHiNGBGSVo9LDCTsI1AyIJzPDLDdw/93MmLFLhxU5SXjE 4DCgTp1zXREqNBqxqIA6VN44gAA1Ix8MCxYQqPoQaGHCECMO8KLHhAQhXrxs0dLGSitZc+e+WZ0T CIIUPJZYqYCn9kk8eQZDfHjUzA8GFBYD6ECcuAIkZCJLrgrW4eTnk5FUkKCHhwcvCIToqJRpyup9 smTWA2EJyAkPfNpEEWAkz1duChk6dDgBAfDFJxLo1+9iQYjkyyGhAh4NTaZcZMkZkAQAeiXhRRFt aKGJPK3QE8yFGXBgRxtspEBDejBgBZYAMAyw0HzPNVTEfRAcEMCLL8aBXGSUTaBCHbsZuAAZPJJh RArseJDEBhKYsQUn3bXlCv8rU9hlBxBlJBFDegZ0JJYZR6iA1G4FqsgiC36E6UcAcUBW1VUCICPW iQQg+B8GSehBgweglSlEaToUwIGFIHDCwShabOGFBDR01oYRRWFlRoliXVmUfEisGBwEYIoZQBEG oLmNAcV0FRZDbfbIEQCeCSGEGRN4IcAlEvbZCidNRiiHEAjQ0EIMJqIpQBQF1GGAASeFxWYNX4o5 ZhG0HfErDECNsowBRiiERJs7EmAADStJgAACXphhBwzwULJJk6xw0IaG28LmhQoMTYBVHlG0oUJQ MBzx6bRxFGtpEQUA5W8UQRSzzEHCgiqZChIsEUMMEnghBGlqqJGJHXZoEU//Jm1s0QatUvKxnlUe 5VFBvwQlU2+0DF2hb5gBIDBbycqoYMmvuIWVUHwYwKCFCTyslMJndqhRmhVCbDGaFnZsYcegSbTD jgrTPpSUEUGQXBIMBuQhAAYRrDymy8cgc2OWBtxsBlbsfQRWHTp40cISEgzZsBqjCFHxFmrYUcAW QkjgIQ9MMIEAZJJxxF4UA0RBUgXKYN1VGF5fOlsyQhmhghGZFcYeN7/WMYAOMSzRThISbNCREASo ESgCSRfgtwQemKCHBAYoR0BERqBUwOKMM+MXDC5Ezq85BuARlhllZ26YACDlgdsRlsjOQwopkL7B oBIQ0LCpQswppR5JtGGG/1UYtAfSWt98w/gyzQQ/aaUsQ+ir1oN1hFSiRX10vxkVbGHCEibwWRLU 0LAhkc4DHliYCZiQBB0M4AjOO0Id/lUJb5AkGTFTgfsWA79jDaB4XzFbo4hyIm1AZzIGUEMMmKAX HtAgbqFBIOxMsMIWnGAKs1GcN0RyjCBUwhnO+FfjNkgpY11qPSMUC1hGFJ9pTUAAB6KREeqQgiUs QSU0iMGcaDAn2QGwDH/y1w4rMZAgeIMZaJwXvYjYwQDUwAhWKczZGqWbaZUvijQyQxTkACQAAEAP erDiHwF4AkvUS1kwUGMFyMG4tWQwiP7SYOQiQqMa3e9RBTJCFKh1oKrkrP8NbXhNAqckJAQ4QwC9 Yc82/jIQZzgykkAxI0HYELkyBYg3lzFKgZB3whS5Kw8qKI0ljjEgDFArOu5iD24MUAAY1GaCBvjL vMyYvgHQ8n1GLEIecJlLpNQPCYeZz3wg8i5uMANaC4FOQ6w0IrFUAJVPhMHmmOUNUJLjmhzMJhJv tr+0fSUs3SCMN0WEB2X96lcQtNkS24Obr6AqZ8a8HR6SIkEdiuSewrscE/u5ld8ZAHEhbFR7zAnE kkJSiOwrHqLMcKgzzdF8y9IhPotoqTiYgTLiRNNH/lKifj2rc4m0aNiCEkRZVrMcKuhKBXKjDTyc rX7sNMAWakmAHvWIRlL/E1E0wZG+grDvpGNrZ0ef8Q17kuMYuytebZo4rRqZwQdUDYFc5wogZFoJ D/GqZzkAY1BghbB+IpKgMwhy1AIMwHFVmhaCeEQjlWGzpgT4D4CWg9NUwqAN6AAK1oz3VLQJwKkC YGpnnxpabjgjCImr16+QsKO50nUB+Xosy+IwgcKxc4n83I0AssFQ90TrRLwxgwpScIL13I9L5Nxp MKMALVS11rWMJQCxZHusIyRzK9HkKcFQ1SYkmKFwVzkbmyaDAQN4IAXG7chTvalMCfZFWovt0QaW U5+MqjIw+KXfBEJ1O8YWzjmLJYAXmlaDmz6kCOzBbzT9GgQ4UsuqtqtK/xzsQ91LWTciRcjwluLQ pg2QQa7JSYBrRxwC/YSAAMRtGnL+swECxAHBdFxIHCqAAR6N2Ko8mm9s81nTOJBhA0D2sI1LvJ8E oAENIVjAHVhwhyY7+chHTgCPUnC7GMj1DlFmcZA9LNcibKDIRoaymEW8Aa89IA4hWDKTnXyANrv5 zQnYQJu/QOcq0NnNd5DyEg7wnwPsIAQH+AKe1cyCQrMADWg+9JrdTOdGH+AOITAzAQ4ghRdY+tJY aEITsMDpS79gByyoQhU8TeoqoIEFWWAyGg6QhSZfWtSi3oGsZX2AOKBhB22W9ag3relNv6AKLDCz BHYwgmKf4dhn6IGye/+A7GKPQArEdra0nS2FO2QhC4zuQQvu0IRpe1sKoZaCuKXQbW87O9jUfcAG pLDsFzUARgF497KXze5527sHI6jCGrKw6S+goAdzqMII7j3vBowgBTsgeA8C0G6GL8HMG8hCGibe gIpb/OIYz3jG8d2FNTCcBD1YwxoGngaNVzwNPUiBFEqe8Ym7vOQPT3fEX05zmjeg5jin+Ajm0AU4 +BwOdBj5zXPu8oQT3eYx53GY1C3xk7vc6Sx/+tAp/vKbNwDfPO9Cz+cwcKdDfegNiEEPbD71qied pktft9dNTvEeWN3kF+/BuMftdrZfPOEWp/jFpX72DqpbCnFvt7Jvju//sV992RUf/OHn7Wxlx3vw C5d3vA+Pd8TL2/JX3wHEm5D4xgv+2SMouRQG7vZiK9v0Ck+9vV80gg1UIQ0BOEO3483wM4wg2ZmH OAlib+kmnIHhA3/2C85w9VEzXO4vYPjoC69wZCc72cwnwbHHdIAq/D72ml74x5sg/R5oXua+38EX sDCmB2ia0yTY/Rno/KJd94AEAu/B8GVfbE1L2/a+l376kf2EM0j/DlhgASTwAA/wBJx2Bg8QAH5g AZzWBAcAcXdGfgHwACTgb1hwbCTQBCiAArv3Al8Aclgwf9x3Bi/QayVof03QAymoaRSIBQ/gBwH4 ghZwByTwBBbwBGFi/wE6+AQvaIMMiG5K5wcPQACrhgLldwYo4G9C+AQbiAVPYIBYIH1ReAYXaIK9 pmn+J4XJZn0w6IILaIQPgAUogIM2+IQ+eINhYoN9cAeSlgBfYAEE+AB94G9w+AAWMH44aAEocAY+ CIMW4AcfyH2aNnzMhoHS9wUr+AVnAIMkgAIE2Ad9cINm6IOTaIYzGDlE2AdCWIBJ2Ac86Ad9gAKS 6IlPEIov+IaMKIUj6Hwk8GspaGfJV4IbqIAbeIOSqIO3OImXmG4EcAd/+IKhiAKe6AelaIsWoIl2 OIYFqINOKIAYqGkZ+Guj1wRVcADddgZ2todfKIplqIOQ+IS5uItBGP8Ak4aDBRiKO6iGf1iKymiH kfiEYXJsITiIJ0gCI/Bp1Sd3PWBqWBAAm8ZpX9AHJBAmkCiMeYiLOsiGvHgAOLiAkJiOx9iQpRiJ YgKOYhiF+leC9igFsTZ8I8B9B4AGxFeN0peEF8iJtZiDCCmOaDcmBPAFZHiMkgiO75iGx0iKOpiA KkhqlkZuxZaBo1aNVcBwbbZwLxBovxeGdKaMahiJCSlpMAmO6WiGD9mQxIiQSWiEL8JsGQhyyiZ7 +Fh9INcEiBh7X3AHX6B9TyBoqEiMkPiOLNlGL2mJU0mV35iGZRgmWHCWB4ACvgeWVYBrblYFKYhv 3KdsL9BkKBAAsLf/lndwAOQngxsIlwoZhEP4BcQolZKYmU05kzI5Ji9CjWypkce2bP7Xaxl4lmh5 BhNXgHyJjOdYi+BYmS1JjphJl+loLBPplMe4gRyYgLCHfb72aijwany5h4ypgFjQZAG5iTA4mbdI m373kpzZmfBYkTfpjaHYaHQohDDygqf5a43mlwsHe6ComqhIgF9Ih5YonUY0l5X4lu9YieqIi2KY hG82Z9w5nhZwfQynh4+GllHonGLYnZm5nGa2AH0QhwmonpO4iQz6gks3gQ56hju4iAkIbzK4nShw khrKhKKYoXH4BA+Ybj6GBmEGZSY2V2AWZmB2ZIUmZjIaZSsqWR+m/x8ySmY3OmZkJjwFEARBAAZK gANE6gI4AKTVpAFKqgFgAAY40KRgsKRKCqVU2qRS+g1sMKUukAMuQKUaIBKGpQE5cAFE0KVBsKQz 1UZFoAE20KYf8KY/EKc/8AE2IANEkAM5gANEsKdEcAEyYAMfEKdv+gEyIAMXcKh8yqc5EAaMCqdz +qYyEAYykANgEARKQAQfEAEIEAE+8AN1eqdcKjwDUKikCqiDKqiFqgQDoAR42qo5EAQ5EKlh8AGY OqiDOquDagOMWqhhQARKcAFcKqR6mqnb4gMXIKlkSqQ4wEbZNKqH6qeNGqhx2qk5sA42cKYDAKs5 EAM+kD5ggABRyv+kWtoGJ9ACZ3qkOKCkYYAAMmACYaABQXABP1CoH+ADYVADF4Ba2YoDoVphReCs ZNqnvGqqj9oCXZAEERADepACQUAETpAENAAA1EMH1WMCCBADAHAFOWACBnsCSYAAHvABLUSxAJCw DJQwsxMEEkAH3ApA+VqmwqMBhhqwfMqrYWADOZAEXXADACABXbAETlCnLeABdMADTkADXVADYtAZ TOABLcAE1eEEPFOugXMFPHCwP9sCNCAGAOAESyAGA3AFHZAETICxehAEkRqzhYqoeFqzhRoEX7sE XWArCOAEV/ABYkADTvADTuABYoAAN1ADsaG0TqBFXRADtyIGYrD/BNsqBifQBVCrtFXbBRrgAk5Q Ay0gJS2gBB8gBzFLpzObA6CqqBfAAzWgAUuALVewBAjwA0uQAhm7BCfQGSZwAgjABDdAXGLwNkrb Ap6xtBKgtUtwBQCQuykQAwiwBF8KQMHLBDVgAxHguf6qAW9ap4aqqK1KBPyqvXvaq4GKqxFQr5n6 AxEQvhEgp3F6vrc6r4WKp0F6pZbasDjwp5kavZ9bvUMaBkrQsK3Kr0cqujXLqLpKpzhwqT/gA+dL vp0qp3TapnaqvQW8v0tqqUowpH86reVrv/4aBI8qA0KiBE2TA/s7pAOQvDmgpEdKwQ1LBD+ALSmA p/Obp4ZawHsK/6RH+qVfiq3wOsIaUD1EIAP1agMIrMHjuKaDKgNdIAYXoHUq6wFhEAQnYAJdoAQ1 IAE4gAC2WwMpIAMDUANzewIakAQpoAQn8MIfUMY4MABxowFyIAEncAUSMABEkAQykAItUAMnQChJ 8AHuALIngMDSW8TU+6YI0AK34gRDywR/ZCt6IAbrIAaIqzBiwARLUAAIALke0gIeUh0ANMkmELF6 gLFMsLRiQLZM0AJX6wFdwAOkogcMNMonUK+BXJtGrKuYe7Vbeys24ASQXAOIXLZ+m7BLoLRh+7AR MLR70LQ5sLQ8cAI34ASocANMUEUeEBuHS0NMEAFd8AF/1AUnEP8EdswDMvADs6ymbCqpNQAASdAC V6DJgRMDKTDJ06wSs1srMTC7QYAAS3tekBwncpJFJwAA67CwPJAEJpACn7wE7HACkJzNLZS3KUDJ RBAGZSCqkjqrP3CzmSoDWGyoRXAFEXABWMyp0hqofxoBNYAAV3wCOIDSOFC9PqACV3AFRMCpNh2+ qPADV5DSmhoBNnACNbDTEy0DzLovA3DRdlqodUqoF2DSQMy+nvqodAqpfvoBh2qnfvrUSS3E0orB 2tvVPiCtP6y9iFrU8QOwayvABIu+bM3WPvDWcN2pQX3Gd2sDrxG+9YvBOu0BEmCongrEMpynoGrW HrS2V02qsur/pqfa1gcc13BNBJ2hBLdyxrN7BSMtA3jcuvrctLARATOtBFx0ASAdBPxK2JcyANZr 2H56rIyq0Y7q1nF9wHEK2SYQBGIAOOqstU6wykmguDcgxD8L2SsgJUd7A95stEkgpqZtxKSa1c/6 rDZ7s6b62oyNA9UxAJAMAE0rtZl7A6jssziLAABwK0xABONdHcirB14rx8tNvYit2s/93NE9q20K qGw9x07QyCcQA07AA9mNubbCBE6QAx+wQNM8t04AtTRwA0a7uEHAqMLDwW6a2vD9rH36wzOL2NJd 3zYwxzFwBR6cAj4gATv9A7br3y2gAUTgIT6gBClwBWIM4i9O/z01AMTlnE0cXL31/d6qfeHxHd88 vr+xaqmvmqc5ELY9gwCs+g2wuqpKeqmwWsFN3bnCAwYLXL2E6sA8Xqh9iqiJ+uVgHuZ7mq6ruqdt 67aRqtjSeuP7AgacKqi22sBbPrNe/uV1LuZ43uXvXd+L/dblzKCX4uacCtdwHucczuG7uuU/vujy Xap/Ot1y6tjlO8sJ4JuiKAMFUMFAuulS+qVH9emgHuqiPurfIKWbvulPyiJQUD0KkAQKUAKwHuuy Puu0Xuu2fuu4nuu1DgUs0hhR8evAHuzCPuzEXuzAvgJO0OvGXuwrYA3N/gcrEO3P7uzRXg3PPu3M Tgfaru3YDv/tyT4pvr7swN4BrMwHCqAAfqTdPMMgfNDuCnADfCAG8P4HHaAAHcAHDFIC597t1QAA YyACAD8G4e7tyi7uwc5CCsAA/67vJcBC5l4CJqDwIgAAE28CJbACNzAGEQ8FvE4FIsDvf8AEAP/q UIARGYHsBW/wv04H2s1C0syzHUAHitwBK6Dd/i0GTtABNxDzuM1CDMIEIC/zJsAHgNMBUYHy4K7y Sr/0xE4HkOBHTEAHR//tizHwTH/1WC/sSF/1Wd/1Xj/1LEINXz/2XU/1jEH2aM/0Ww8BRpvzxfH2 cB/3cj/3dF/3dn/3AEAFwaEPVCDxAP/3gB/4gj/4hF/4hn8k+IQ/BgyQGBYBARTQ9wwQ+ZI/+ZRf +ZZ/+Zif+ZqP+XoPAYEAADs= ------=_NextPart_000_0000_01CB3ACA.1C64E450 Content-Type: image/jpeg Content-Transfer-Encoding: base64 Content-Location: http://www.stata.com/distrib/policies/webforms/juul3_thumb.jpg /9j/4AAQSkZJRgABAgAAZABkAAD/7AARRHVja3kAAQAEAAAAPAAA/+ED4mh0dHA6Ly9ucy5hZG9i ZS5jb20veGFwLzEuMC8APD94cGFja2V0IGJlZ2luPSLvu78iIGlkPSJXNU0wTXBDZWhpSHpyZVN6 TlRjemtjOWQiPz4gPHg6eG1wbWV0YSB4bWxuczp4PSJhZG9iZTpuczptZXRhLyIgeDp4bXB0az0i QWRvYmUgWE1QIENvcmUgNS4wLWMwNjAgNjEuMTM0Nzc3LCAyMDEwLzAyLzEyLTE3OjMyOjAwICAg ICAgICAiPiA8cmRmOlJERiB4bWxuczpyZGY9Imh0dHA6Ly93d3cudzMub3JnLzE5OTkvMDIvMjIt cmRmLXN5bnRheC1ucyMiPiA8cmRmOkRlc2NyaXB0aW9uIHJkZjphYm91dD0iIiB4bWxuczp4bXBS aWdodHM9Imh0dHA6Ly9ucy5hZG9iZS5jb20veGFwLzEuMC9yaWdodHMvIiB4bWxuczp4bXBNTT0i aHR0cDovL25zLmFkb2JlLmNvbS94YXAvMS4wL21tLyIgeG1sbnM6c3RSZWY9Imh0dHA6Ly9ucy5h ZG9iZS5jb20veGFwLzEuMC9zVHlwZS9SZXNvdXJjZVJlZiMiIHhtbG5zOnhtcD0iaHR0cDovL25z LmFkb2JlLmNvbS94YXAvMS4wLyIgeG1wUmlnaHRzOk1hcmtlZD0iRmFsc2UiIHhtcE1NOk9yaWdp bmFsRG9jdW1lbnRJRD0iYWRvYmU6ZG9jaWQ6cGhvdG9zaG9wOjVmY2RmYThmLTMwYjYtMTFkOS05 OTI1LWRiMjI3N2YxOGRkZSIgeG1wTU06RG9jdW1lbnRJRD0ieG1wLmRpZDpBNzZGREYxQjkzNjUx MURGOERGNUZGODk3NTcxREIwOSIgeG1wTU06SW5zdGFuY2VJRD0ieG1wLmlpZDpBNzZGREYxQTkz NjUxMURGOERGNUZGODk3NTcxREIwOSIgeG1wOkNyZWF0b3JUb29sPSJBZG9iZSBQaG90b3Nob3Ag Q1M1IE1hY2ludG9zaCI+IDx4bXBNTTpEZXJpdmVkRnJvbSBzdFJlZjppbnN0YW5jZUlEPSJ4bXAu aWlkOkZEN0YxMTc0MDcyMDY4MTE4RjYyRTgxMjQ5NzdGODk2IiBzdFJlZjpkb2N1bWVudElEPSJh ZG9iZTpkb2NpZDpwaG90b3Nob3A6NWZjZGZhOGYtMzBiNi0xMWQ5LTk5MjUtZGIyMjc3ZjE4ZGRl Ii8+IDwvcmRmOkRlc2NyaXB0aW9uPiA8L3JkZjpSREY+IDwveDp4bXBtZXRhPiA8P3hwYWNrZXQg ZW5kPSJyIj8+/+4ADkFkb2JlAGTAAAAAAf/bAIQABgQEBAUEBgUFBgkGBQYJCwgGBggLDAoKCwoK DBAMDAwMDAwQDA4PEA8ODBMTFBQTExwbGxscHx8fHx8fHx8fHwEHBwcNDA0YEBAYGhURFRofHx8f Hx8fHx8fHx8fHx8fHx8fHx8fHx8fHx8fHx8fHx8fHx8fHx8fHx8fHx8fHx8f/8AAEQgAaQBUAwER AAIRAQMRAf/EAKwAAAICAwEBAAAAAAAAAAAAAAAHBQYBAwQCCAEBAAIDAQAAAAAAAAAAAAAAAAEE AwUGAhAAAQMDAgMEAwsJBgcAAAAAAQIDBAARBRIGIRMHMSIUCEEydFFhcdFigpLCFjZGsUKyI3M0 FTcYgZGTJFUXodIzQ0SURxEAAgECAwQIBAUFAAAAAAAAAAECEQMhEgQxMgUGQVGBwSJyEzNhkYI0 cbHR8ULhUmIjFP/aAAwDAQACEQMRAD8Ak/M11T6h7R3fjoW2Z7kWI/D5ryEtJcBWFkXuUm3CvUUe WX/y47o3Fu7p4rLblkLk5ETn2Q4oco8tAQUjSnSPzjUMlDS8K18r6SvjqCQ8K18r6SvjoA8K18r6 SvjoA8K18r6SvjoA8K18r6SvjoA8K18r6SvjoA8K18r6SvjoDXJYbbjurTq1IQpQ7yu0C/u0BW/t PJ+zHjf+/p06vlWtf++gE/5jXnkbthBDikAxeISSPzveqhq34kdlyzFO1Oq/l3F+8uy1r6fKK1FR 8dI4qJJ7Ee7WbSbnaavmJJanD+1DPqyaEKAKAKAKAKAKA0zP3N/9mv8ARNALv8E/OoBc+ZD73QfZ frVr9ZvI7Tln2p+buLx5dchATsswDJaE5UuQ6mKVpDpb7o1hF9Wnh22rNpH4TWcx25evmp4cqxGu laFX0qCtJ0qsb2I9Bq0c+0ZuKEHLkctisa2hzIzGITbitKFyHENBSvcBWRc1DaW0yW7U54RTf4HQ HWjostJ5gu3YjvC17j3ak8UZ6oQaJs+DBjKlTZDUWMj13nlpbQL+6pRAqG6HqEJSdIqrMsTYUiO3 JYfbdjvW5LyFpUhd+zSoGx/spUShJOjWKCZ+5v8A7Nf6JqTyLv8ABPzqAXPmQ+90H2X61a/WbyO0 5Z9qfm7hZRcnlMVMxWRxDqmcmyHeQ4gBSwVLUngkg3uFEWtWCEmqUNvqLUZ5lJJxwrUfXSGVh9qY l5rOZNLO6886Zb+LkugLStV+UkpP/TW6CCQo34irtmkVi/Ezl+KW7monmtx/0wwzJfPsQoGc31C3 RvtD6ZUr+MJk6g0lS20Q0oX3tY4JabbSDq1f21WUpSl8TeShYsWGvDkaw/y/VjE8y8fEzsdgcqzq luJW6wicy4FxUiwUpKrak61HsNx2Gs+qpRM1HL6kpTjLDY6PaVzpc/Nn5rGbi3dkXIeA2y0GsZLd Vym1OCyUR0cP1g03KrX4Dia8WsWnLYi7xCNLcrVmKdy4/Eljh1vqJDr3v3K5DIQIOGnkbbfjh9p+ Ms6JThUUrBUnieXpto9B7RU6i426LYYuDaKNuLc1S6njXoX9es1ZrHSldCGY24y+/lIs0SoMNTv6 +NHc7ra5DRuvl2JIChw1Dso01bx2i1OMtc3CityVG6YNrqfWVXYbGXzzmPxGSmOxtm4d8zZUkKDT MaxKrh21ta1cEp4njwrFbTlRfxRe1co2lKUUnemqLpb7OpDL66dSXpG24De08kF4yY641kZ8ZZuF IAKWFKFlI1hRPH1rVYv3cFTYafhXDskperGk0vCn+fxLN/8APfmfVq0c10i+8yH3ug+y/WqhrN5H Z8s+1PzdxKdPt2N7R6LSM+IAnSmci6zHSRwSpzSApa7XSgen+6vVmeW3X4mDiWl/6NcoVosif7fE U8Hbe8N5ZRc3HY+TKenPqLsoJUWA6tV1FTyu6nTe/bwrAoSm8Ebi7qrNhLNJRyrd6fkNbrJJ24nI 4vBZbdEmG23DSxnIMFkuKeUUgpddUFIBJKR3VX4dlWr1KpNmi4VCbhKcYR8T8Lk6diGP0r2Ptvbm 00s4mYrLQMkRL8S6ElCwpICdLY7qRYfDftrNagksMamo4jqrl2741lccKCA6k7vy2/dz+BgYt5De NLseDBYStbtkqOta2kjgo242HAVSuzc3gjq9DpoaWz4pb1G5dH7F32xhshtHo9mF7knr27IyT+rF rUjXIZUUpTqS2DqStzSb2IIHGs0IuMHXCprdRejqNXH016igsep/sbOi2y9j5TPq3TCz8zK5iH3p seQgNKU68gpU4q5WpbauNuPw1NmEW61qzHxbU3oQ9NxjGEtjWOzo+BCdc96ystnlbJgY5TMXHPiy GwQ4++pHApaSOKe93eHHtrHfuVeVIucG0sbVv1pPeW3qX6nf0f2vmNt4Pc2c3KheHxLsPkt+LRpU XQSUu8pfHuE2TccSa92IONW8CtxXUQvShbt+OSdcPyr+ZeP/AJ96x9X1/T6vrVbOX6RfeZD73QfZ frVQ1m8js+Wfan5u4meisPAYrZk3eealuohxQ/EkRlXcjBrmIWVllIUVrKiB2Vk0scKlHmLUSV30 18HXpG39q9ow4bqmp8ZMaG4yxIQwpKgwuQsIbC0ovo1KV6RVw5ptvafOvU/ZMTPb9mZTA5/GSmJu iQ7zpjbXK1JII1K7q0/qVHuEkekVTu2JOVV0nTaHi1mNqMZ1UoKmC2jTgY3ZLuzMdsJ7dKUPY9uP KkyYUhLWu6y7pDnEaVE8E3vaxrOrXhUTUy4g/XleUV4q7S0t766dsZFbbOThCUtvmyJKFItoGrvO PDu9rShxV21loUHJtUKV1zx2E3ltRBxuch+IxUtpKkc5JbLkohpDa1I1aFKURpKuHbesN625LA2f C9bGxJqa8Ml8ipdNto43bOHnK3FmkxlbnjPQYLmKcMhTbLB1SH1PMpWhOglI970+9jtWGq16S3r+ LQlKHprMoOuK2v8AAa+G3P01xkHFx2szFeW1FS1GmyHErfLTTIXd10jUk8s371qsRikjS3b0pybf S606Do6mbYO8NiTcbDkpQt9CJEV4G7ay2Q4kEpv3Vgdo+GvNyGZUM2g1SsXczVVRr5le5avsBy/z tOi3v6bVkKnSL3zIfe6D7L9atfrN5HZ8s+1PzdxbOi23MXuLpWvGZRC3YKsm466yhamwvlFC0pUU kHTqAuL1n0m52mq5j+5+lHHuJ3p1s/IZLauTgTzjZEWE+/PQ+l1a9L7zie6SlQDahxPEqv2cL1ao c/Ui87O6a4KVj3JGIyiIeVxrUwPc0FbLcvxCkEoTrQl79V3O/wAb/JNKCpF5DK9JW1pbx+Myss3a MhhcltgJS6m6xxK/XWo81JIN+zhSgqdORyWxV7bwmaGLyTi3XxAdbdmBK4vhmg+lxC0IcQoNtvjQ VaU29alBU6o28eisTCy8cxBy7cLIpYOiyQtzw8oraS1dy9yp0qPye2lBU6Ubm6Sy4cbDMx8xJZxk tbg0rZAcdecbSlTjinEJspS0hJBSQAQbXIKgqQjOb6QGO63Ni51qY7zOaGnUPEsMgtNnmXSChwWG kjgeB7tKDMOrZm5MNl9uy4eKYkRmsHfGOR5egOpLDQAvoUv830+moJIL8E/OoBc+ZD73QfZfrVr9 ZvI7Tln2p+buGB5c/wCXqvb5H5EVn0m52mq5j+5+lDFl4fETHEuy4MeS6kpUlbrSFqBRcJIKgezW q3wmrJoDD2Gw76Qh+DHdQlAaSlbSFANpBAQAR6oCiAPfNAaBtjbQCwMTCAcAS4PDtd5IGkA93iLc KA3fwTDcoM+AjcoEkN8lGkFSQkm1rcUgD4KA0Hau2CCDh4RBBSR4ZriFK1ker6V974eNAevsztvl ut/wmHy3+L6PDtWXdQX3xp73eAPH00B6VtzbylJUrFxCpF9BLDRI1Cxt3fSKA9qgQIcOZ4SM1H54 cde5SEo1uKTxWrSBqUfdNAUL8E/OoBc+ZD73QfZfrVr9ZvI7Tln2p+buGB5c/wCXqvb5H5EVn0m5 2mq5j+5+lDQqyaAKAKAKAKAgM9jt3vzm38JlWYccISl2PIZDqSQSSpNtKgbW9NARcnC9UF8vkbgh tHSnnExgq6gDfSLcASakgmcZH3GzjZgzsuPLeUFFlUZpTSUo0dhCiq5vUElP/BPzqAXPmQ+90H2X 61a/WbyO05Z9qfm7hgeXP+Xqvb5H5EVn0m52mq5j+5+lDQqyaAKAKAKAKAKAKA0zP3N/9mv9E0Au /wAE/OoBc+ZD73QfZfrVr9ZvI7Tln2p+buGB5c/5eq9vkfkRWfSbnaarmP7n6UNCrJoAoAoAoAoA oAoDTM/c3/2a/wBE0Au/wT86gEv5t9zZHE73xbMVLKkuQtai62Fm+sjgTXl2Yz2ot6fiN6wmrcqJ /gLDb3mH6o7dx/8AD8NOjRIZWp3lCIyrvrtqN1hR42r3C1GKojDqdXcvSzXHWRJ/1V9bP9Xj/wDp Rv8Akr1lK+YynzU9blGycswpXuCFHJ/4IplGYmcB5hOuWWU+F7ig45LASdcuEyhKtRPqkNnstShN STkdaut7bsRDe8cO+mW4pBcRHY0tBCVKKnDy+A7tuHuiooKkZn/MH10xDjKE7hgZAPBR1w4jDgSU nsV+rFiRxFTQVIZXmp63pNlZVhJ7bGFHB4/CimUjMTW1fMN1qz8t+Orc0GAWWw4lT0Jg6ze2lOlH bRxJqcm4vMj1oxmUexf2giTEJCUreahMJSdY4jii/C9MoqfR/MV/t7zb9/Rrv7+m/wCWvB6KR5je i+e3vuSFloMtlhiNG5BbWhxayrUVE93hapToQ0KD+mDdn+oM/wCC7XrMRlD+mDdn+oM/4LtMwynb hvLzv7DZFrJYzLNR5rGoNPCOtRGtJQrgoEcUqIqMwyk5/td1l5qHftGjW2LN/wCV4Cx1cBpt28aZ iaGhjpD1cYbDbW4UJaSoqDfhjpuTc8NPZf0UzCh0/wC2XWflcr7RtaNWoDwg4KuVXvov2kmlRQgc x5ed+5iaZuRyrL0pSUoU54dabhAsLhIApmIynGPLFu5Pq5FofAy8KnMMpn+mLdxUFHItE3HEsu+i ozDKfT/8Hn/YXwfL/wAxo06ePbp015PReZ3YKA4qAKAKAKAKAKAKAKAzQHZ/41Af/9k= ------=_NextPart_000_0000_01CB3ACA.1C64E450 Content-Type: image/jpeg Content-Transfer-Encoding: base64 Content-Location: http://www.stata.com/distrib/policies/webforms/isp_small.jpg /9j/4AAQSkZJRgABAgAAZABkAAD/7AARRHVja3kAAQAEAAAAPAAA/+4ADkFkb2JlAGTAAAAAAf/b AIQABgQEBAUEBgUFBgkGBQYJCwgGBggLDAoKCwoKDBAMDAwMDAwQDA4PEA8ODBMTFBQTExwbGxsc Hx8fHx8fHx8fHwEHBwcNDA0YEBAYGhURFRofHx8fHx8fHx8fHx8fHx8fHx8fHx8fHx8fHx8fHx8f Hx8fHx8fHx8fHx8fHx8fHx8f/8AAEQgAlgB4AwERAAIRAQMRAf/EAK8AAAEEAwEAAAAAAAAAAAAA AAIAAQMHBAUGCAEBAAIDAQAAAAAAAAAAAAAAAAECAwQGBRAAAQMCAwQECQYKCAQHAAAAARECAwAE IRIFMUEGB1EiExRhcZEystIjczWBsZMkFVWhwdFScjOUFhcI8KLCwzSENlbhQoIl8ZLyQ4OzdREB AAECAggGAgIDAQAAAAAAAAERAjFRIYHREgMTMxRBkbFSBAXwcWFioeEyFf/aAAwDAQACEQMRAD8A uLnzxzrnBnBkWraNI2O7deRwuLmteCx0cjiEcHDawVMQiZedXfzUc0A0nvEP0MPqVbdU3peubfUb h9nDK53WfE17j4SwOqjIKS/na4DtBtkB/wCl7wPwNoEL+dGK8dZfRBFA7r27BaWtc9pAJLQDipXf QGy8uCxXhzCoAzIFwcqeQUBRXUzjif6IT+KgPvEqoWuad2bDd+WgnicXnElN3kH5aCXKOk+WgWUd J8tAsvhPloFlHSfLQai711tvq7bBwTMxrmuXEkkrQcpzv4Us+KOFItNu7iW2iFy2btIQ1zlax7UR +CdatX5fyuRZvUqrfgoI/wAvnDJBH2xf/RwfkrzP/f8A6f5/0xVepII2stYoxiGMDATvDWJXuRNY qznla0v3OIc/HLl85zzsO4L+PYKkA9oWLDYvojwUDSMa4hVzdmUQkfnHcKCcRRtDihVVBJcfz6Ar fOVIA8p/NPgoJmiVxUjMSf8AmOPy4UGXGx7QSgLvHQH1+gUCV64geWgQL02BfHQJZF80J0r/AMKD h+Iy799LT3Q+eg2PMH4TH7wei6vK+46Ovapfgrk7K5VhWs39U39H+zXd23aIbVBvhy3MyJkL1GVC VIBI/SXH8J3Cr1QTo2uQ4bOqApwK49J8G9x8FKhp4HMeAiLGVG1B1sD4aiokyI15DTtRP/N4amoV uxWo4J4PkNKjNhgbHiR18cVJ+epEyigVAqBUCoOF4k/1pae6/HQbHmD8Jj94PRdXlfcdHXtUvwVy dlcqwrXYjoWgHa3d+jXaW3aIbdGS9jTIS4ktGfMNyZ2k/grLN2n8zQGCKaOUZlcXBQSS8guJClwG PjqYmYD3DJFJeccpCAqUxA3b1qbggUzKSi4/1k3VFQmMGXapRdx3dFTAywpQZjjj+DxVkQFpcHbS QFVf/CgkEoKncPAaB84w8OOygQe047h4DQIPBOG7wUHD8Sf60tPdfjoNjzB+Ex+8HouryvuOjr2q X4K5OyuVYVuQQx9pCcoUoT5FrsuHbpht+DMuI0PaDD849BGx34j4Kz8S3x/P2iETX9iPN9o4HKCB 1QDsUbW9FRbNI/kBmVcwK7yfJu/puFISJoY3a3HYQTgh/p8lXwQOPKpGVBvKr89IQmD8AAD0bquC Dmkn8tSECNi0D5mnFaBZm9NAg4fJQcLxIV40tPdfjoNlzB+Ex+8HouryvuOjr2qX4K5OyuVYVqR3 M7MhAacgCK7wdCV11t0w3PBkfacx6rmR47QprPzrp8IRSDtvC8BvZxlrQjduGCfNU70z4QJBNmam Rg2YjaEwG2rRM/wgTRGesqg7V3/gq8ISNaxjipXDeN1WgSAs29HgqwRdECCUXA7KBNdEpRFG3CgY Oi80Jh4KAw1hGAw8VA+UdFBwvEYA40tEG2LHy0Gy5g/CY/eD0XV5X3HR17VL8FcnZXKsK05L7SLc iO4vI4pQ0ZmOGI6rT/aFdtbZbTFtQE6tobwAy+iccxCAFVAK4eBKvu25iJ+uaGzM03sRxyqC7aCW p5eiomgkbrmjNaC68iQ9UFXY5kwGOPnCrRMIolbxDoalzb2JAM5xcoBVMP8Apq29BQY4j0UAg30W ZpARTgpRPHU70FBHiLRmsY517EGvbmBxQtwK/wBYU34zKEOItGJawX0Re4Fw2ogUnyAU34KCZxFo uV575GezdkkOKBygJs24034zKHPEmgiJ0pvY+zaQHOUkKU9YU34KHfxFojH5HXjA8DMWlVAxxOGH mmm/GZQmcR6G8jLeRlccDuUNU9AU034KON1bUrG/4vtZrOZs0YiALm7FKHf4DUxMSN1zB+Ex+8Ho ury/uOjr2qX4K6aAXAHYTXKsKi+K+a/MKLibWLdmsO7GG9uIo2ugtnIyKUtYFdEScrWNAXortbfj 2Uw9WK75F8TixrzmNzZtLOzv7y/mgs9RZI6xuZLW1DJmMdkk7Nxi6wDihqe3sy9SfkcTMFtzO5p3 cNzcW2pyTRWMfeLuVltakRRl4Z2jj2WAzvAXpNO3sy9SPkcSfEEPNHmhJZzXcOpyPs7MsFxO22tT HEZijMzhEjc5Zh0pTt7MvU7jiZoDze5hIV1rDafq9pu3/qvDTt7MvVHc35sm+5nc0rC4bDfanJb3 BZHO1kltah2SZgkjkHsv+djw4HoNO2sy9Uz8jiR4pdO5ic2tULu4X8t0Gvihe5ltaZWvmOWFrnGI NbmLEap3U7azL1I4/EnxYR5u8xWueDrRDsWv9hab+qQfZfJTt7MvVHc35si35o807iG4uLfVJZYb NgmupWW1sWxMc9sYe8iLqgvc1q9NO2sy9U9xxM2O3m7zEyhrdZOQKjRb2qBUX/2vAKdvZl6o7m/M R5vcxyq6045vOW3tcVVV9l4TTtrMvU7m/M38XeYyr9suUgAnu9rsBUD9V0407azL1O5vzWfyE4q4 g4g1y/frF4bt1v3cQqyOMNzdoHYRtYMcgrLw+Hbbg2ODxLrsV/cwfhMfvB6Lq877jo69rJfgrpvn Dx1yrC8vcYr+9uupt+0LtPp313duENS7GVoP5q8Bs062tbSC8hm0W3dBpFxI1sjXOk0h1m5zYyB2 P1oRyYly4uw2VarNzLfz9NZpPMfhW24/4h1yeOY6bqzbARsbA3M50F3aTz541DRmFvIfCfHREcSN 6ZbzSedHDTHx9/musn1SS6aLZj2zG21C6myOaCAfq08QBP5uXdU1Wjiww9F5s8Ls0vSxqvbS37Hx OugLWJzYLpjrwy6gxwLQ90neojkAHmJubUIjiQyLrmrwHcufcXneL6eOKSItltGA3Zl0y0tCXOzp GO1tXlMfOXpocy38/Sey5w8Jwa1eSTahe3GnXN3aXeR1lE1nYQX00/cuzYQ5wZBMxoc8kKECNSlU xxYq155rcJw6PDHaCZl7FYyw2TRbMAsZ/s9tqkb1OcTXI7YuTq+OlUcyKMDSOZXDNlxpxhqT2THQ telgmt7JsLUlEN7DcPikjXK3tImSM6OtjREcSKzk5fmRxHpmva1aXNhLJdd3s2W9zfyxCB9xK2WR /aGMEojJGsx/NopxLqy5SjGVBdH8svxrVf8ALf3tTDc+NhL0pzB+Ex+8Houry/uOjr2s9+Cum+cP HXKsLy7xkU4s10jD/uF3j/8AM+u7twhqXYyt13LnhO4sraf7Dlt7ZljdBzQ6eG/77FpPfGtmjkL4 p2uc0yxywuTEMc2rM/LimH5RhnkvoljaXLZzfXt/Np113a3DYxJBe2tzaQucjHBjv8X5mYgYq7oU RyoYdpyw0LTOaHDeiXLn6lpmq291cTw3PVyut+9R5XOgykgPtg45fErtpI5cRdEA/cLgHiF1lf6L d3FvBfXDrHsLSFxt3T2WnNvLx0HenduBI7MyFjt5C0Ny2cGQzkdpbop5BqF8il8JEEZbCzsbKYRX ZXqTHv6AD8w9OCieVA7Tkfod9cXUVtqN/Gy31B1hHNPFD7UQX8FhO9rGOcfPnLmlxbsAQqSFDkwF nJXhqeGOaDVNQbH3eK8m7SGBeylivXZWZXefn047cEd4KUOVA5eTug20F1NbXlxcN7rIou4WKx0l tZXcT4zG9vWay9ykuUKFTGlDlQmg5C6JPfiKHUtQkgNyLHKy3idM2T7RnsHzuAKCFvd853hcTShy YUxI3JK9irkcWr4ilGuGguj+WX41qv8Alv72phufGwl6U5g/CY/eD0XV5f3HR17We/BXTfOHjrlW F5e4w/1drn/6F3/9767u3CGpdjLD+19W9h9euPqrSy19tJ7JrghbH1uoCNwqyKyd2s6w8hz9QuXO awxNJnkJEbgGlgV3mkNAIoVkMuqanNcNuZr24kuWAhk75ZHSAFVAeTmCqd9Cso4ru7h7PsZ5Iuyf 2sWR7m5JMBnahGV2G0Y0KtracZ8RWmm3unwXSR6iXG7uHNa65f2mUvHblZEd2bVx+c0Wi+WvZq+r MYWMv7lrHPMrmtmkAMhIJeQHecSAV20VrIRqWogILycBA1BK/wA0ZkG3Z13eU9NCpfaWpEEG8nQh CO1fsQN6fzWtHiA6KFZbLQ+M+ItEuX3Njcg3DkLZp2NnfG4PMgfE6TMWOznMo2nErRMXzDSkkkkl ScST00VNQXT/ACy/GtV/y397Uw3PjYS9KcwfhMfvB6Lq8v7jo69rPfgrlUxrlWF02p8oOG7u5lvI tD0d3bkyudPbhz3OeGl7nuyHHNnNdpb8eZj/AKlsbtuUMRnJrh4uA+wtDeFJ/wAM0E7co/Vp0Kan t5913mbtuUC/gloZd8D0MNLlP1VuDcxKDqbcqCnbz7rvM3bcoSHkpw6WtTQdEa5RmPdmkIMuzqb+ tTt5913mbtuUBHJPQcrv+xaGSQA091aEOKu8zxIKdvPuu8zdtygjyT0DMU0LQ8pdh9Vaoav6GJp2 8+67zN23KDu5J8P5GhuhaIHBoDnG1ahcjVKZMEOZKdvPuu8zdtygw5KcPh6nQtDLACA3urRmchyk 9TDrIqbvwu3n3XeZu25QX8FuGwJAdD0Qlz/ZnuzeqxQU8zEotRyJ913mbtuUIzyY4fMDm/YWhicl Q4WoLQBlwTJivWqvIn3XeZu25QkfyY4ZLyWaForWAI1ptWkl3Wxcez8WypjgT7rjdtygzOTPDQPX 0LRHADdagKcw29TYG9FOR/a43bcoYOl8I6fw3xPHDa2lnavnY10vcYhC12UIMwAbsJdWfhWU8ZlN IjCKO25g/CY/eD0XV5/3HR17VL8FcnYa5VhXXC1YI/0B81d7ZhDZP2TdxOPhNWDCFo3uPymgbsno cfFify0AuDg0Yqm1FoGJcGHpd49lAI7VuIOPQVK1AlZIXNxwO+oqBcXbqrKUTg89PkNUmoEumGwf 1TVZqAc+cf8AoNVmbkuN1dzzxnb5ivshuI3+GsvBmZRLdcwfhMfvB6Lq0PuOjr2sd+CuTsNcqwrs g/UR/ot+au9swhsjqwF0jWlCQKBnPZsJGO6gB3ZOQAigYtYXEKCcQcBtNA7oSoK7NyDZVZAubISU VPEDUSkiS0Iqu3u+XZVZkCsjjlDipXwbdtRpkEZHNeATmLAjkODt5w3EbaTdMSUSSkOiJ2gonlq9 +m0hwGsADjS3QJ7IekapwfFMt1zB+Ex+8HourQ+46OvaxX4K5Ow1yrCuyD9RH+i35q72zCGyOrAC AXeH8FAGQqUOG5SaB8qDDymgJqptC0AzZuriN+01WYEYc4FPKniPRVUmdgp6CU+TN+SokE2RrGez a5zlCuyuQodyDyVMaMBEx2doRc2Ae/ZiuG1CoqkQGLnPhyqWglpw3Zm5k8tKViiXD6ov75wAlfZD 5zU8KKVRLfcwfhMfvB6Lq0PuOjr2sd+CuCMDXKVYV2QfqI/0R81d9ZhDZHVgDgzN1gNm2gYtJJxQ JsTdQNlbsankWgdrUG35U8lAMylACVCkphuNRIiC50Q7/mNRQKRUdhvd/bpRJdq8RNjaCDirtm/p +c7qeAhRHK0EEYZgMekAD5h8pqKB4yHRZmg4PDU3ANagTpw30oOJ1Rf30g90PSNTbBLTfzP6rqmm cvoLnTbuayuDfRsM1vI6N+UxSkjM0gooFXiInFS7B5K/iBx3/uLUv2ub1qnlW5R5KVS/xJ5i/wC6 NW/bbj16vQqX8SeYv+6dW/bbj16UKl/EnmL/ALo1b9tuPXpQq63SZOaGpW1rO3jjUIGXMcUp7a7u mACZxaAxxeGy5QOvl83EHZUJ0smSHmVDbT3M/H2oNt4JXRGSO4u5AcrJXh36xoR3YlEJ2+JQnu9M 5oWz4on8w7kTzOcIoZL26jc5sbQ57+s/YC4AL5ytI85KaDSintOZ8QsiePL7Ne39tp+V13ctLDdR teJj7Q+zbnQ7KGkOgWnNbXIJ57Pje+ZHDqP2cTLeXLQW9n2veVEhAjykb6aDSn1HSebFhot5qdzx 1eMNnb95ltDeXYkIyuOVvXQlWp+Gmg0uB/iRzD/3Pqv7bcevU0RV2HBs3MHiXTp7w8e39k6CUxmC S8lLiEb1xmnjwJcmyolMNVxdxJx/w9q32fHxrqOoezEj5ob2bKHEuGXqyyBequ3fRErB/lz1/XdY 1zU5NX1G51F8XdhE66mfMWh3arlLy5FSq3L2rJ/mze1vLe2VVdqMTQgJx7GbopaXPHGPQfIavVSh vkPkNKlC+Q+Q0qUL5D5DSpQ+Z+GLkGA24UqULM9EVybUxpUoWZ21XL8vipUocvkO0uKbFXdSpQ2Z 5CK5OjHxUqUIuedpcfGtKlDfIfIaVKMyx1nVrD/A3Uttjm9mS0qQm1OigivL68vZzcXkr553edK9 S4p0mhRdX8rgP2xq5Qp9V3e9qly9r1lrbbV1m4XL8kaFShPzVVZWl1Z8DGdxfqDQ5cRkk9Wgi7lw F94N+jl9SgXcuAvvBv0cvqUC7lwF94N+jl9SgXcuAvvBv0cvqUC7lwF94N+jl9SgXcuAvvBv0cvq UC7lwF94N+jl9SgXcuAvvBv0cvqUC7lwF94N+jl9SgXcuAvvBv0cvqUC7lwF94N+jl9SgXcuAvvB v0cvqUG74ZtuE2XbTZXjZJFwAY8fO0UH/9k= ------=_NextPart_000_0000_01CB3ACA.1C64E450 Content-Type: image/jpeg Content-Transfer-Encoding: base64 Content-Location: http://www.stata.com/distrib/policies/webforms/saus3_thumb.jpg /9j/4AAQSkZJRgABAgAAZABkAAD/7AARRHVja3kAAQAEAAAAPAAA/+ED4mh0dHA6Ly9ucy5hZG9i ZS5jb20veGFwLzEuMC8APD94cGFja2V0IGJlZ2luPSLvu78iIGlkPSJXNU0wTXBDZWhpSHpyZVN6 TlRjemtjOWQiPz4gPHg6eG1wbWV0YSB4bWxuczp4PSJhZG9iZTpuczptZXRhLyIgeDp4bXB0az0i QWRvYmUgWE1QIENvcmUgNS4wLWMwNjAgNjEuMTM0Nzc3LCAyMDEwLzAyLzEyLTE3OjMyOjAwICAg ICAgICAiPiA8cmRmOlJERiB4bWxuczpyZGY9Imh0dHA6Ly93d3cudzMub3JnLzE5OTkvMDIvMjIt cmRmLXN5bnRheC1ucyMiPiA8cmRmOkRlc2NyaXB0aW9uIHJkZjphYm91dD0iIiB4bWxuczp4bXBS aWdodHM9Imh0dHA6Ly9ucy5hZG9iZS5jb20veGFwLzEuMC9yaWdodHMvIiB4bWxuczp4bXBNTT0i aHR0cDovL25zLmFkb2JlLmNvbS94YXAvMS4wL21tLyIgeG1sbnM6c3RSZWY9Imh0dHA6Ly9ucy5h ZG9iZS5jb20veGFwLzEuMC9zVHlwZS9SZXNvdXJjZVJlZiMiIHhtbG5zOnhtcD0iaHR0cDovL25z LmFkb2JlLmNvbS94YXAvMS4wLyIgeG1wUmlnaHRzOk1hcmtlZD0iRmFsc2UiIHhtcE1NOk9yaWdp bmFsRG9jdW1lbnRJRD0iYWRvYmU6ZG9jaWQ6cGhvdG9zaG9wOjVmY2RmYThmLTMwYjYtMTFkOS05 OTI1LWRiMjI3N2YxOGRkZSIgeG1wTU06RG9jdW1lbnRJRD0ieG1wLmRpZDo1MjY3QTg4NTkzNEUx MURGQjQ1Q0M3QkQxOUJFRTY5MiIgeG1wTU06SW5zdGFuY2VJRD0ieG1wLmlpZDo1MjY3QTg4NDkz NEUxMURGQjQ1Q0M3QkQxOUJFRTY5MiIgeG1wOkNyZWF0b3JUb29sPSJBZG9iZSBQaG90b3Nob3Ag Q1M1IE1hY2ludG9zaCI+IDx4bXBNTTpEZXJpdmVkRnJvbSBzdFJlZjppbnN0YW5jZUlEPSJ4bXAu aWlkOkZCN0YxMTc0MDcyMDY4MTE4RjYyREI1Mjk0OTdBNjVCIiBzdFJlZjpkb2N1bWVudElEPSJh ZG9iZTpkb2NpZDpwaG90b3Nob3A6NWZjZGZhOGYtMzBiNi0xMWQ5LTk5MjUtZGIyMjc3ZjE4ZGRl Ii8+IDwvcmRmOkRlc2NyaXB0aW9uPiA8L3JkZjpSREY+IDwveDp4bXBtZXRhPiA8P3hwYWNrZXQg ZW5kPSJyIj8+/+4ADkFkb2JlAGTAAAAAAf/bAIQABgQEBAUEBgUFBgkGBQYJCwgGBggLDAoKCwoK DBAMDAwMDAwQDA4PEA8ODBMTFBQTExwbGxscHx8fHx8fHx8fHwEHBwcNDA0YEBAYGhURFRofHx8f Hx8fHx8fHx8fHx8fHx8fHx8fHx8fHx8fHx8fHx8fHx8fHx8fHx8fHx8fHx8f/8AAEQgAaQBUAwER AAIRAQMRAf/EAJMAAAICAwEBAAAAAAAAAAAAAAUGAwQAAgcBCAEAAwEBAAAAAAAAAAAAAAAAAAEC AwQQAAIBAgQEAgUHCQcFAAAAAAECAxEEABIFBiExEwdBIlFhsTIUcYGRUnQ2CKFCcsIjFRZGF2KC ktJTJhjB8TMkRBEAAgIDAQACAwEBAAAAAAAAAAERAiExAxJBE1FhIjJx/9oADAMBAAIRAxEAPwCf 8TndLuFtXflpp+3NUns7KSwjleKJQymQySAtyPgBiqpEuTon4cN27n3R25Op6/ePd6h8fPD1pRlb poqFRwpwFTiLTOCjp89yIULPIVp48xz4DEWtHyAv6lqWvXIdLctalHHTKEVIU/nE+nHJbvaRZJIH 3C6BxcSHLzBWgOGrdGAdjuyY46h+o3BlpXjTwx0q+EM3LuTQsUb0f98VMgbv1gVCqxrzoRSnrJ5Y HIHsjFQvECp48amn5MUwKl7NNCC6SHgCWB4geU5T/ipXEOZAC/xXcfw5+8Kftfd+flXGgCd3Xd13 HGAaD4dPa2Obs8nL2eRi7YzkbbowzE3EvE1oOC+jFc3g04v+Qjr2oOLm2s0YkS5pJOXBYx/mIxn3 s3hGrZDYOkswOYZq+76vScYUrkEyG+3LS6SzEgiq1PRUfWGNL9LPAnYYNIRpYhJK2dkaimtaj18M bck2slBGaJJIyi+8vu0NKHnjZqQKkpMa5pHy0ZVINaceHh6fDE2YEgty2aQkoAOCEeFPHDj5AyaF lspQQr+Rga+ih9OGkBz/APkan9v9bFADe7H3jj+zp7WxzdtnL32Fu31zFFtlgyyOxml4KOFDQcTU YKWSRfH/ACDdU1i5h15pZLaVLUwZEcglRRq8xWmM25LdoZBpG8Y2v+naRm4nAPIEqP0mPDDU1yJX ksWm39Pl1N9TuYzNeyNn6jsxCmlPKPDE+ioH7S5OlaUWSgHHIPlx1c3/ACUEoUuQA2ZSzCr1BAJ8 KYrIyX9ueYT8uHkDCszcCVA8eZwZAjn42s/9lHH5DgA53/I39/8AWwwBvdj7xx/Z09rY5u2zl77C mw1nO16oFyfESjMxoMxKgCvz4dE3UrloBb13zBt34jTy6XOvOFS3tI+KAvU1lc8FCcyPHDXMvIna Lqxt86yXhudRvH6t1JEPfkY+4iryUeAwrUkybe4gK2G+pYr57dSJWiOWRc6lQ31Xc1UHw4VOI+op WfyPmzd4DUNcl0O7gNpPc2vVgUESANE+VxnFOPm5U8Ma81iDStpHq61jTrWUQPLnuDyt4gZJP8C1 I+fGrcFuyRSutyRW65pxFZrzzXk8cJp+iC7Yn2L0Dj3C2+DQ6xp1fVKxH00wemHosJuqxurab4W4 s7vMjGlvcqWPl+q+XB6YvQq1/wBiZqfnVp48+WLLB3dj7xx/Z09rY5u2zl77AdxucaFsuArV5bm7 kjS3zFQ4pUsSCKZKYrnqBUTaEnT9DuNQvJdQvn6sk5aWV3kCHjU1Z2rlqeQ5nwxbuau8LBcuIumY 7LSS8ZjzNJfUYNI3Py08yRgcAobjzOF7RH2fJFp22NXtdMZxeXEBgZrnp26hYpJSOEhlkOVaGjEv QjC+xMatIe23t6TRQNd16c2AuIvh9P06yZ2urhHOeQgsA69aTzFiM1KcueB3Gxy0213frEbLbsu3 dHJAdIeE75vGSdszOx8aV+XDrkFIasu3e2rVurJbyX05ArPN5iWr9aTO/HGiqPyaXku1bPUVspLO zhEQzXvXusjwh1PTDKA1SSOVeAOblghD8/osansjbt5aTF9O6UoQkOuVyOB8ynnQ+o4PKDyCekv9 Pejxy06dfGnu4ZqUe7H3jj+zp7WxzdtnL32LItNIu9J6V88hlQEwRInUqGkoQq/WPMmo4DGatAqP BJdWenT28LxhhEVAWKQCMk0APnXn5R5aDlyxDsxuoInt57WeaOciKKUAQB8zPQ0OXl48gAPnw/Uk OoR25qc02jR6zq9qfh4ZOnpmlZTnurgMRHLKhrwFPKpNK+Y4ppJwjZJIfNtbWeef9+a7It1qtx5g jMDHEta9NQeSj8vjjalPyUkE9zpr1pqem3miWsV8/Tlgu85JyRsUZAq9RI+DAmpFaCgPHjuWkgpo GpaxPp9g+s2qR6jmlE6RsqIoDlVlCsxOUp4e9xwpGDte2jtG43Ha31zaB59SaSG5nSV4/P0KBvI6 0LxRGM05rwOCQkY72W3jiRI8gRUdAwZQFXJTLStTXAEiJ/I3qz/rYYwb3Y+8cf2dPa2Obts5e+wH YlW0k2vW6T3YdIo+H7R1cEBl95lWpNB44wsw5rA2ag+4NJ7eCPb8L3m45JUhjYQxtIj1BkdxKVTK EUjzHxAx18UoN1oqy7o3xe6Rqt1DoptLpGtH0uNLeTrPE9yY7hHWTNUiJKkqBQNjSEMo2e7e6Umq 1fQ4TAkEnTthazqmaKOWUkSFzR3KxxAMeZqMEIDaPuP3Ha7jmfQT8LHARJAllc/tWd41jm45nSoJ pF7y8c3hioQSb3G9e6ssllNHoZiBkdLrTY7OUNnWWLLG1xIWXjFnbOoy0NPeGHCALaFu3uHqU7xX Ohx2sD2dw1vd/DTxk3CKxhfJK3lRqAdNvN68KEAR7ebi3bqy3EW4dLFqbaG2KXBt5LcvK6t1KiTy tXKr+SmXNlPEYGhocJ1iNrKSqghG4UHoOJGc9/kb+/8ArYYA3ux944/s6e1sc3bZy99mmzdJgmSx 1N3CtazSAqcoBFa8zx8eQxl+CuSwNe/m1q32rLLt4TG7+IjJjtohNK0R4sq+WXLm4DNkaniKVI7a pGwrR7t7lhrlodvlDZwzsRcRSh5WhzFYYmVwnUlWlGFVzcBghBLNV3d3Xs4J1OgdcnPLC4ikdV8y qIqKytxZ6g8TQNzwQgyWr7XO5S6Rol5b6XcyXvxt1da1DHCmb4KOYpFBRqGrLKrLl8xCV9OGoCWQ T727xN0LRtC6Ej9Mz30dtIEQMUct52dcqjMj1+Uep4CWZB3D7n6laW95ZaEGt7iJLiFoIpXXy3Bj ZZHJYFWRcwCDNl9HiQEs6BtvVNXv9v2V3qtsbTUJoybm2KGPKwYgeVizCoAIrx+TliLMZdlSQ20x p5Sjew4lIBEp/sWn9v8A640GDe7H3jj+zp7WxzdtnL32S7UQrt2GTLVZbiWMg1ClloV8w91uPCvD GNlgvl/kL6zqG47fakVzoqTyXUd5SVIUEs3RyupohBJo+StPZjo4P+cmouQ7g7wPSL9zVkjzRPM0 DdJvJGRIVEi53zBxVSF48K42hCybRbx35q23tcWCzax1TTrq2jSKzt2aQ5mczw52M1aKijqKhy14 rghDk8/ibvFkksotMmzZpJHu3tFMkSBw0ca5H6Tt01ynhzPD1OEEsfdi6xuPU9Ouptwaf+7rlLqS O2iKNGXt8qsjsrM/GrMvPwwYGg3Lbxhi9u/Rl8cvun9JeWEMhF8I5Ft7mLLIx4upqtD48cECLlyF FpNl5dNqf4ThjOc/yN/f/WwADe7H3jj+zp7WxzdtnL32TbMs7jU9vT6bbZhIXPVk45IxK1A6jMql 1CE8sQqtorkv5Nb3VtY0i6i022vDcWYYgXyZZEcItCuY8M5NAyjiMS3amJKbgprurdfQus2pdN0K tG4jRiq8eAXKK1I8cT91ifbCdnu3WJbg/EXFUHBVqjCnvZyQooxHhhfdZfJasMWhatPPqLwmZxal OtmqA1WoAKZflx0cujsxplvcmt2+k2q3UmoNaxO2RXkBKlqFqEqrZfKpNTwxrdWj+djYBm3iIJhn 1aFuDkqJIySIkaRyAoJoqIWr6Mc09ExSExvPbTCJJtViErqnmZwP/IoccCvDymtDy8cdfxkcgC93 JqTXTNo+qrNp5YpLCTHIoFONGQVHDjTHG+zVv0TmTeo/gOteGaub5+eO41B3dj7xx/Z09rY5u2zl 77Ab6Lufce0V27pOqzaRaXjz/EzW8RZpGNAqPMCpjTnWn5eWHyt8FcXgO9u9JOibVO19acz6dpKu ttqF0nRWbOxpGkR81Y/Bganng9JzOipTFu9k06C4jGgszXU+aK5trtXESIWIVi5OYgmtOGMIWzPB S0htzvqT2RtSY4eErJ+0ARfE04mo9WFalWsB/wAG6x1XW7e8V5YKLwEakUYqW8a08MSm6uSvTGtr NdbtBa3kJuYD52Ryy0LAr4U8GOOvlazLMt+2W0bUK66aruFlUtmcVFwnTkU0bxQ09XhjaqhbK8i1 re1NoSTvYw6LSbNVxmkPEUofe9Vccvbq24WyGyS32joWgpHa6bbmFpEMk6AMFMhWgrmJxHSrlTlg WKH+AqU41pTwrXHcbHIvxSdxNybc3/a2WmNbiCSwjlYSwrI2YySDmfkwfWrbM71TeTmGm/iT7q6Z bi3sry0ihDmQJ8HC3nIoTVgTyw1zSBQtA/We+vcLWbv4vVJ7W6nCBEZ7ZKKo+oo8qn1gVw3zTBwF NH7p7/1YXNxNqOn2YQIpmubQnq5eSq6K/mThzPiMZvhX8C8IYbPuf3Msbxmt9w6UjxjMtwbBjmVg WzKGizEBhTlwwV41WkNVSLad9u7t3e3FpNujSenAsTrPJp4aNhKGJpSIkZcvGvzYt1TKPJ++nd+2 BNrubTZ5VVy0MWngUyKWA4xgeY0UesjDgC4nfDvG6Avu3TF4VYHTmOXlzpD9PowQElbSu9Hdu81u 5srXc2kdSKLrPd/BRZWbOV6YqobNwrTC8rYkkBt0/iN7w6fqFzpcus2F6sagPNFZQ5SWWpAJWtV5 YfmQZ9C/Fz/0i+NqPiPg/iM1OHU6OetP0sSWJX4j+ym6t57ut9c0y4tktYrOO26UnV6mdHdifKrL TzjxxScCaOQ/8Zd/f69r9E3+TD9C8mf8Zd/f69r9E3+TB6DyELDsN3b0+3a3sdWS2t2YyNDG86oX IAzZclK8BxwvQ4LH9F+9efqfv9i+UJmM1yTlFSF4ry8x+nBIQQp2J7vJdteJrCrdMojaYPOGKAEB T5OVCcHoIN4+x/eSOVpo9cySuCHkEtwGIJUmpy+JRfoweggkXsz3tVSq7gcKSSQJrmhJFD+b4+OC QgFSfhu7jSTSTPdwNNKxaSQmfMzE1JJyceOD0KDRvwzdwHJzXFsSeFT1j+pg9B5PqD+G9S/ph+58 v/t/CfD+NM3S6dfTSuJKOh33IYAKWADMAGYAMwAZgAzABmADMAHuAC5/82AD/9k= ------=_NextPart_000_0000_01CB3ACA.1C64E450 Content-Type: image/gif Content-Transfer-Encoding: base64 Content-Location: http://www.stata.com/distrib/policies/webforms/ml3_thumb.gif R0lGODlhVABpAMQAADo+QwJjqtHMoV2jzPn6+NbW1bi4t5KUb0STwrGyhS+JwYmMjaPK4uPj4vPz 7Bt/vPHrz9Xo8sPFw7y9kujx+Hm11+Du9r3a6p6fn6WlfGZpaevr6xAaIQp1sufarP///yH5BAAA AAAALAAAAABUAGkAAAX/4CeOZGmeaKqubOu+cCyLRG3feK7vfO/7JcICQCwCML+kculpOp9OiKM2 IgA42CxnMeu6CNCwczOlOTTarMbLVoHF4QLZfE1v23jTG+4USBoONBt1aWt5hx97fB5+coIaGkZE GIiHinwCBo6JGwMPDwqhCg8VlXmXcJmbBBsIHa+wAaWmbahiqhuCrrCvsixUtCi2Ybi6vL2zKQ5N EDENSDLDUMWcu7y+KssezQQFEt4FiRIGBB8SGwUOBRoEDhLf5x8NDQUNH97vEomLfZq51cc6YFPG TJ6kbpEoDdGADoADA0bWaFgw8UMkAJFy8WtCjZW1WMlQaGvW4IhJdUO4/zS4I8EhRAwT1yygyAUA TQD2NjLyZ+zYQJEFS2KwCYDAUJsfMHDQ4M6lTUgSI9VMifODzo6tAv48MfKDgwUSMEhYgBDsB7AL 5pEtgMFAW0oSJuoTS3fKVZ4AfYbkWjCYnrursur9QCGChQ8XLkRIlLivXxLSnmD9iOwDAlADBCpg gCDAAwbbPmxYsCFsUn0lCpA1YOCsvRUOMGyI3C8wZYGlFAQI+EpBBQkkORQwQNbmnw9uCRjgYBQJ Rn1GyZnDYE91VcD/PGot1bkDgwu9I3j6TRIA26KQcB4Vi35NkWdENGFkh/H6xsnbLXu2QEHggA8M kAKcQWEBYNFSFmEEVv97H5i0zkwLDNXSIBDldB9e2g3W2QOG+fdBBQIGZ50VRJynAWvMQdLgEGMl 1BIkDtRnIT/4aehZhwH8ByJ5Xsk2nHLDSSDEAg48g1xrGLC1wWhkJbVAOMRhYNeFtuW3IY46hvhY ENj1dA13N1rg4Y4DJiKCBd/hcAEDEZRTzggWrMnAm2iWc4EFbnaZ15f6cShmjh9qacEAFRQ6wKGE GppoBQN8V2ihCCBQgQWGVjoAnnpmyKduHeCIQKAVGNDMBZImioCpplYgaQURKLAqo6wSeuijCFxg FZXZCcYnrBRQwCgDiFVwQTqEfWcsA9/JKWeyFxDALLIXUGCssBdUwAD/BbRxhKGuIG05Q7Y7VTnY C+3cEEgiDqSLB7g18vmCAxDEKy+88tZ7bjSZclsZuVKkmy4B/fob75v44uplt97GwO62t22V8C/5 NrzXwyssLK67XtTQK7bRbIytxblKrDEB2JZM8skmoInAKJ+AosAA0a6g8icdPFAzAvHSyDBvkYri s88PDECweAIJtFsASCdNCgUnDJr000kzDTAmOwd0m1YIvFlB0bz1YrQCi5EQINRk3zs1MVUfc1nX vGRNwAC7sR330xyOwADZZD/AdBUQoH1x2zaz/cqnFcQtONdHK8D0BXjj/U8VfofMc+CCI8DA4V57 nfS1DzQOtd4maFOb/+RWU842zZhrXnQArnpO9uMkTJM24DWnbnssRrue994l9D36wbBcRrPNpt/u k+ZI56477COIHi7paitQM/G1U2887sjrjjTooUs2e/Ast0z9+DVTW23xWq2OtOWJMaDb6yco0m7b 4Ys/fPXAknAB+scbXQHBonmf0ngHGe/9DXz2a9n0ape/EmwNc0j7jwkYB7V7FfB3e6JfAhPYAQUA cAQRONzcbKUHAXqGgLHDoKY0uEHxdWBiZ+KfT7hnArg9zYKQU6G+Ble/FnoHBRaQ4TU+hQIbbg+F IzAg9FjYQpo1sARBhCART2DE3TCPBkoEHg9B0USbPZEEURThFGtIN/8kiiCLGQTfKHrowi/CyXpd C8AYS1DFB1xxHzq82spcxsVP1M+NMaxd5VJQRzPeSlsHHNxlQsHF8LEMkB+IohAFMkcS1PGO8vue In/WSPFBUpKC5Bkhy4gCNK4weCvj5BofCcT7xbGSI7hk/Ey5ww707Gei4OMnFQjHto1ygKXMI89S ictcfgaI0nNl8WApAln+RZhWI2YxGblL+4UyeKM0Gg1LQEs9SnOax2xa+Ba4zF9uD5PdHKarpknN Vm7wmsz8ANx6sc0UInKJqFwnO8OZMja60pbZPOIsoam2lRkUnJ/0pyslSMXknTCY99SiLSPVs2+2 U5xNFGQ85ynQZ0b/NI2vGIBBD2rRhPbxnR1gKBmPVs8kErRtIqVoRX9m0i6GNKAP9ejzJHoqmfqU optB5hozqlI6zq2lZ3xp8ETK1J/2jIQmoAAxT+rColryqCisgfPmBwtEefWrPYVqCSjgs0h5FahW jaXhFJeIdjhgSRtIZ0Bk9dW6wgwFFEjUBPY6gQQkIAMZOMABNHCAf32wAq6aiWzeCle4yvUYi3qU pRh1V8hoVQCYFQBf/frXwB4gAxuYx5KmUIMJfFawoGVsY+Oq1JBK9rWvtZUN/NUORmCWr3vlLGAn QAF/+fYhgkUtIFa7pMe2DSoQihAGMJAAvjbArYatQWY1i9vc+lUC/7/1rWmDewAyEJe1Hz3l4KDC 3Qw0V7OaDRgOHHDb6vZVtxPIrr8MwN3uqraxxoXFAt7r1wlMt70Q2AF7/btZ/vo1A/GVrwO2i1rv Eje/IaWue/kqgG3kYAoD3ixnD7xb6P6WwYJ18GohnNK/bpjCTwiwubSK2w13trMJli+I7ftdEg+A s3vFbBSioNV/YVjDLgbsbhW84PKKGL+t7cAC0KvjbUwjwKT17Qbea10OCznG2QVxamuc5AHY9haZ hfJvWdHXKndWyAgm8oyP7Nguf7kP/52AmH8LAQOfGc1pVvCa79vm8NbSy/8N9F4D9tsNBPnAZz4A lj9sZD4XN8kVYP8ERwJ922Zwxc4bNu9uaWAmKghAyJ8NxIV9BwGsDECfqVyAlfGMZgPk49Xv8Gx9 Zy0WWOcDA/V1tath3eJtocEOwA62sIdNbGETgQOEADYAtpXsYjt7KTOJxERSMm1IPHsLy23LAozN 7Gtfez7STg9ULnLtCEHo18rutreLHZF2R2IpRXC2gtLD7cA0e93APhFNYCKVai93CPImArmDvWx7 45vY987CFRZ+7IPXO1cJdzgWTsTvmUBEuRhHt8TTUHCIb1zZUAG3wMfd8I9roeOCiLjJV45vlHNC 4yyPect9LYma2/zmOM+5znducw18IztFGke2h070ohv96EhPutJDk+5qQDQPHe9gjdSnTvWqW/3q WM+61qv+DjmY7a30KIDYx072spv97GhPu9rXbvYGkAGAbv2u3OdO97rb/e5wNWwIAAA7 ------=_NextPart_000_0000_01CB3ACA.1C64E450 Content-Type: image/jpeg Content-Transfer-Encoding: base64 Content-Location: http://www.stata.com/distrib/policies/webforms/mais_thumb.jpg /9j/4AAQSkZJRgABAgAAZABkAAD/7AARRHVja3kAAQAEAAAAPAAA/+4ADkFkb2JlAGTAAAAAAf/b AIQABgQEBAUEBgUFBgkGBQYJCwgGBggLDAoKCwoKDBAMDAwMDAwQDA4PEA8ODBMTFBQTExwbGxsc Hx8fHx8fHx8fHwEHBwcNDA0YEBAYGhURFRofHx8fHx8fHx8fHx8fHx8fHx8fHx8fHx8fHx8fHx8f Hx8fHx8fHx8fHx8fHx8fHx8f/8AAEQgAaQBUAwERAAIRAQMRAf/EAJ4AAAEFAQEAAAAAAAAAAAAA AAUAAwQGBwIBAQEBAQEBAQAAAAAAAAAAAAAAAQIDBAUQAAEDAgMDBggLBwMFAAAAAAECAwQAESES BTETBkFRIhQ0B2GxMlJywxUIcYGRstJTc0WFFhihwUIjhEYX8WLCkjNDRCYRAQEAAgECAwcFAAAA AAAAAAABEQISMQNRYaHwIXGR8SITQYGxBBT/2gAMAwEAAhEDEQA/AD3vUd5PHPCGr6Ezw1qLsJqV HeXIS2kKClJWAkm4Na1wzcifu08fcY8VcL6xM4hnuTJMeellhbgCSlsshRAsB/FSrGv9ck+eagXX JPnmgXXJPnmgXXJPnmgXXJPnmg865J880V71yT55ojzrknzzQA/zjM9i7/8A8+93Obw5rXqKmcYB JksXSD0DtAPL4asSmdDCRGcsABvMbAD+Ec1Bn02f3vxNcnuxo70rTW5EpUNC20OJLLriUtDInKte 5S2pSBfHPy1FF9C1jvJOtdTn6dvol5K3JDzfVmh/Nuylt5JdukN4AFFz+2qB0LXu9povF/R3Ftqe dkpS42h1e4dKd0wgoW2ElrpdFWOFiagYZ4p71W3lMuwEiZMaW+zHcjqUW0xmUZyhtKr2dWvAE4Kv 8FAS0nWO8l5eoMztNdZaciTHoiy0ErRICLsIDmdScVYJQL+HnqhuTqXeW/wq+t2K/H1Rt+MWEQ2g l5bRC962SS7bpJTdwJNr4py3oIh1zvaQCwYEhQZdLjr/AFRtS1IEsENNlKwhwFg5SRbZcc9QXDg7 UuI5+nLd1+IYctJsloslm6cyrKxWsElITcC1vDeqAv8Ab/8AW+sqC38X9pZ9A+OrEpnROyufaf8A EUBAkgEgXIGA2XPNQVDrHeinTns0SAqc0UbjdrBDqbLz5gSAMclgMdtFNDU+85TBkdT0xtsDMsl5 JSnKBdBVvMCbnNc4ftITBL4/VGQ4hmEuSpbpDCHEbooAbCCVFWYjy/JN72vhQMIkd6Bjup6tpynU g7p5KwbqBxSoZkpHNe1A71nvIWwQY2nMygtlTaN4VBTedQezdLDo5bEUBHRnuLly1e148RmJlVkL CiXM4PRv0lCxFAZ5KIo39v8A9b6yoq38X9pZ9A+OrEpnROyufaf8RQEKD0EggjaMaCqRu7TheLv9 wl9AkNPMuJ3t05X1BSyEkZc2GBtRcvY/dvw3HlMymjJDrDqHm/52F21BQT5N8pUkFXPy0Mu3O73Q XZDkh12W448suvXfIC1EkkqAAG1VBK0Dg3RtClPSYBfLr7aWlb50uJShJBCUgjDZQHKIXJQUb+3/ AOt9ZUVb+L+0s+gfHViVUdZmcQRWGFaXqRgNKUsPBOmnUStXRy4hbeSwvXXt6y9f5w493e6/TKIz rXeFHhzXw/7akMqZQ1DOnHT7hy5U4lZU7nCU7cMK6Tt6Z9/u/dy27u8mZM34LozNkdSiyFtrVnZL j6Sndu58xSEJbP7zsxrz7TFenS5maMKjNBaAFGymypShYgKA2HZXLa7Z8sO0kwYKAlxppV8zjC3C sBRSlxNsDbYMdm01ib7enh7fJeM9fF2zHu4hLigEbnM4oEghy+yxwt8d6ab7WyXw8DbWSZZrpHFG rLlqVJ4ralMIQsiOdHVFzqGCRvi4q3S8FfQva8vV8/8A0efoIaXxNxJJeCZiWI7ZISFMfz1ZjsGU JTTbtazoad7e9RWfrD0CFIlSJTyzHTmLEeEp51XgQhKrk47K5zSX6uvO+0Vf2l/8F7SyPf8Ad3+T cnf7c1txtz/7K549+HXP25X7i/tLPoHx1ItS+D+ySPtB80UpB+oqBPnOM3KOi2hQQpWFyoi+F8MK x3O5NJmta623EdRJhWlBcIKHElSHLW8nygrk+OrrtNpmdEssuK5XrEdOO7WWvrABb4bba3hnL2Zq IaSzucqy/i0TsUcLAW570U3113MBvk7wr3YRYZM3NfyqxynismU9h0OtJcAtfak8hGBHy1pHZ2UG Xfdv4j62gs/F/aWfQPjqxKl8H9kkfaD5opSD5IAuTYc5qKHT4a3kLCOm24oLum10qAtsJAII8Ncf 7HZ/Jpdejfb347ZRlxpDUQNLTlQELQ0u+OdZ/jtsB2VrsdvhpNeuDub8trUNLs6RHLbUZSrYE2sL jkua7ZcsJ0UIb0tuM9cSk4obI6QXmzJy35ucVGnpiOCy1NFLqXd+VFQ3YNtl73tfHZXO9uX55WWz PmIQt1uAltYctfMoecTc+OtofOygy37t/EfW0Fo4v7Sz6B8dWJUvg/skj7QfNFKQ/PfQqepl82bb SChPIbi96sSmNKk5dVLDJJZcSoqTyAjYaVYMS+nkY2B4kL9EC5Hx7KypuWpTJZDbqW0BQG5sLqHM nEVYlcs3lRRdC23UXLbiwAcwJsRif9KWEuTqk9chAZshcSDccitvjqK5gQlxkuFxzeuuqzLVawwF hYUEo7KDLfu38R9bQWji/tLPoHx1YlS+D+ySPtB80UpBSfpcabYuXS4nBLiDY25qmVMsQoelpzNh Tjzpy5lEZjy2vgAKo7XLacjtTU4BpZCknbj0VAc5qDyQHZSWltIBSlQUly6FXA5r7KqOusOstJaL dnVAhpOYG6vg5hy1FRZDhS6mFvChtpCcxBsVEirEpvTpjiNRETeFxpYNrm+UpF8DSkGzsqKy37t/ EfW0Fo4v7Sz6B8dWJUvg/skj7QfNFKQfqKHzEJlqWy5gy2QMMFFVr3vyWvQcRozTTzbJupnKoMp8 1W1Xwkg7aBkQ2oLpK33HCsWQy30MByqsf21eqO0vQXW3EtoUiUmxGY9PMTYEKxwuaCLJZEmXu3c+ +8jrQICf+kYVvj9ucz4Of5Pu44vx/RL0XS22B1lay5IUCkkiwTY2IA+KsWuoqdlQZb92/iPraC0c X9pZ9A+OrEqXwf2SR9oPmilIP1FRXo7wdU4zZQXYrQTbEC1waD2PHd3u+esCkENtpxAvtJPPQDdW cdiTesKSVMuJACuQEchqxLETTm3NQnF3KUsJSoKcGAuRYAeOrSQW6kvLk6uzm2b3G3w5bfvrKprD KWWUtJxCRa55fDQdnZQZb92/iPraCje9N3mcXcG6xobOgvsstzI7q3w6y28SpCwBYrBtWtYztWLR fef74oiFIj6lGbSo3UBDYxOzlTWuKchrhz3kO+vW9S6l+YoEH+WpzfyYkdKOjbo3y7TepdSUbkd9 nfO1AkSk8aaK6qMFlTKI8e5Kbmyehjmth4aYXJs9+HfX7VVARxho60IbK3JQjx90k7wtpRcoxUcp Vh/DjTBlLf73e+xp3cK444fU4VEDKiKpspBtmz5LYjpWPJTBlFmd9XffEmIiji/RHmy+xHU8w1FU hIfbK8/k3yoPQV/uNqYMudO78++eWyhb3GGkxHV7w7p2LHukNuFslRCbDycw8FMGU4973fUnU9Og njfRHBqCXViQzHYdbaDSUkB3Ki6SrNYCpgyD8Ze8B32cMzWIh4lgzlPNb1SmYLICMbBKsyNtWamW j/mfWP8ABH5o3qfbHVPaG+yJy9YybzNktltm5NlZVL95fuh4j46naRO0qRHaa09hxp1t7eZ1KWsK BTkSoWsOWrLgsyw39MnHv18X5HvoVeacSPuyceHa/F+R76FOZxL9MfHn10X5HfoU5nEv0x8efXRf kd+hTmcS/THx59dF+R36FORxL9MfHn10X5HfoU5HEv0x8efXRfkd+hTmcXqfdl4+SbpkRknnAeHi RTkcSV7s/eArypMdR5L74+NFOZxbv/j7XP8ACP5U6PtDqXU95ZW7z7vLm2ZsvxVnLTbZvkVBCoFQ KgVAqBUCoFQKgkf+vQf/2Q== ------=_NextPart_000_0000_01CB3ACA.1C64E450 Content-Type: image/jpeg Content-Transfer-Encoding: base64 Content-Location: http://www.stata.com/distrib/policies/webforms/musr_thumb.jpg /9j/4AAQSkZJRgABAgAAZABkAAD/7AARRHVja3kAAQAEAAAAPAAA/+4ADkFkb2JlAGTAAAAAAf/b AIQABgQEBAUEBgUFBgkGBQYJCwgGBggLDAoKCwoKDBAMDAwMDAwQDA4PEA8ODBMTFBQTExwbGxsc Hx8fHx8fHx8fHwEHBwcNDA0YEBAYGhURFRofHx8fHx8fHx8fHx8fHx8fHx8fHx8fHx8fHx8fHx8f Hx8fHx8fHx8fHx8fHx8fHx8f/8AAEQgAaQBUAwERAAIRAQMRAf/EAKoAAAEEAwEAAAAAAAAAAAAA AAYABAUHAQIDCAEAAgMBAAAAAAAAAAAAAAAAAAQBAgMFEAACAQIEAwMGBwoNBQAAAAABAgMRBAAS BQYhEwcxIhRBUTJCFQhhcYGydBY2kaHBUsLSI1MXGLFykjMkRFSElFVGJ0fRYkODJhEAAgECAwcE AQMFAQAAAAAAAAECEQMhEgQxQVFxMhMFYYEiMxTwoUKRscHhYiP/2gAMAwEAAhEDEQA/AHXvM9WO oW0t92unbd1KS0spLCOZ4kQMOYZJATUg+QDFlQq2GnQHfW7Ny9P/AGnrt49zfeOni5rqFPLQIVFA B2Zji8YphUsj2ne/rDi+RBUXtO9/WHBkQVF7Tvf1hwZEFRe0739YcGRBUaapugaXZtd3t0IYVIXM xC1Y9gFaccVlFJVo3yNLduU3RHeDWbieGOeGfmQyqHjdSCGVhUEEYtkRR1TozFzrk1tbyXFxcCKC FS8kjEBVUeUk4jIuBMU5Oi2gvvDd+vXGxrnVdrXLTXSlTG0OWQ0V8sgpxHd8uNNPCMmsypXiNWIR hdy3cI/qhM/W69+oPtWp8Z4Suf1ubyvS+PNxwtv9xfDN6VKi942eaPe1sEkZR4KPgpI9d8dTQpOD 5nZ8bFODqt4UdEHkk2Nmdi7eMnFWNT6uM9UqTEfJJK7hwLAA7fgFcLiBimAAL0feOo3vUfVNuPCi 2llE7K4Y5u4UymnZ3s/HGjsZVnq8d3AfvaaMdPGe+QaUxmIAf1Ns9O1jaN5BJfR2aWk0byTuCyq6 1GQgEelmxbT6iMZN8MBvS3pWJqVK1WwnLSSx0Tblqs0wNtY2afpB2skMQJZVFfIK4rKVZU3y2GNJ XJ4La/7kZ7V0rd+ytSkt5xDazxz28s7ggRsg4tRqVA4HFoOUGnNUfA2cJae8qqrjiQ+iWOn6b0mk t1u3uoWt7q5E9mTFJIMzSNkPEr2UJxa5LuSqt4XZu/dq1StNpJ50/ZXzOOTw+aleNOXXt8+EqOtN 5hl+VPUAfeQ+3Ft9Cj+e+OroOh8zteM+t8wq6GU+ov8AfZ/ycZ6rr9hHyf2+yLCFAhPnIH4cLHPI m93No1lrNpo9xOqX15TkxE8Tmrk8nrFTgpKuCw48DaOnlKDmtiNI7Oe11rVNWlWA2htRkWKMi5/R d5s8hNGrTgMWq3huM0q0VX/gg9H6p7futGbVrotawpdeGXmd3MQokqKZvRU8cRK1diqNZpf88B67 4+ankWOFQL2nr2maZabsvNNkMl0UiaNbx1WNpBLKzcviOGTsGNb+bBTwTdFl248R/UWXOVtS2ens Ndz7h3FPougbpmiD6fPaTW9xAXbkmUySRrmWoNWTjjSOli04VeaOyW8tprdtTna/lWq5IbxbmubX ZEWi7WS5F7BN4m/Xlcwjngih4GgVwoH38ErPcm+6ll/jjtoHaXcz3mvksN36wLKsNyaTf3eq7aKN NqNvZv4wOuUSMqBJULUyhcz0AAxg4zTq18dxyrunlCEbm57P8Hfl/wC2XK5Y/msnKrw9GmWvm8lc KVxFqutSvfeQ+3Ft9Cj+e+OroOh8zt+M+t8wq6GH/wCF7P67P+TjPVdfsI+T+32RYTGioKDsr93C xzwK3BoOnXXUTb+pyzSrdwoxSBImaJhFnZS8gBVeLdhONo3qRceI1b1Uo2pW0lle8kN5b20rbmnm S6Qzc5zbvGik0DIWYHiO1RwxhBSuNq3RyjtqW0ukldeHCoBbj0Hbth0/g8dZ3OkWvtFp44Y2SeRu ZEQhZqsACi4ajduybcUs3r/ob0V+9O66tSllx3DXXtlaZeaZ4u2jTQIru8ie3lvZA4uI54+7lAc5 OJrlOLQ1LTebkVteQnbbU/nuw3DzaT6tuqxvUkS2hl0aRLazlkjkeJRIpEq8kOqcQtR5RjK5Y7bq m23jiV19iNrKlXFHTbCaVf7i3PZxQyaQ9r+lvL61l7ziGdiylXVwFPbw8nDE3Lc6JuVU9i4F9ZCX at1eauxE1s9tuanuPcOvaXHO0lxCYpbpzW3Mj0d1QcDWgUnhijuTcVGSy0F9U7sUrc6Ujs4k9T/b +mE94mVv7yH24tvoUfz3x1dB0Pmdzxn1vmFXQyn1FpSpN7P+TjPVdfsI+T+32Ce43ho67uXbNT49 kqBRqVCZwK9noCuF1Ce2nw4+ov8AjS7Xc3Av1D3ne6brGk2GjzwtcG5jS7tSC8lXZCvAeTKTX48W haU3SSlhinu5Dej00ZW5Smt2BX+7YLzTN06jpepweL026vGvZMhLyCAM7DK/ApQHiMNq2pJTi6TS p6V9ToWLiuWk7bpNfH3wDzWtc29urp41xNZzTRxkyeDiYJKogZkMg73oKvE4UtSlCeWq7i28MTmq 1ds3qQaTeFeY80nd+2rza9vq15lgsYJjaWsNwAqhrcDvLUtmIX1j2YrONyCpJZpb8vqRd0k1ccFt 2kTvC83do02nDTp0uodU1DmZobaiqCYxEGoXzd0njwrjWFmM+v8AjitxbQ27TjPNg0uIZWWiGHV9 dea1tYYL8ZEltwwmZaMGMtQAW71eFcZ55YV3CPceGLdOO7kQewLDRtD2/LZ28sxa8nuCgu4zBKSq hR3GCmlF4GmC9qFKS9TfU3Z3nnkkqcCX/wBAYV3ixW/vIfbi2+hR/PfHV0HQ+Z3PGfW+YW9B1J2Q GpULe3BPyBTjPV9fsI+T+32RJS7dsz1Kg1g2934sWbVnyjwlcvLAzfj5SeGMu98cov8Akz7Xbwy/ uBnULSXsd232v6beE6tBbRztYCEuQhyQls3Edne8+NYSjcj2pbGPaO+pqNqa+Ce2vuPHguvqeNda OKSeTTmuNbsDCFSaR1yh5nqHDZKd0fLjONtKeWLov35mULa/ISWHywYK7QurXTBf7m1SOJNHv7ae 0isYCxoZ3IK5B6CkRNQ1xvdjNRUIfKcabd50NQ+5JWo1zwxbOm4ptJOi6JtQ2EulQXTLfxTM4mKC 6ZgGIGYtVKeXEQc8zuN4LBr/AGZaeVyMZ3pPNLpe7YWzuxU0/RoL6S7mtbTSmjllEKI/NWPKqowY Gi1p2UwrSUunaci1b7k1GmLZJaBrFtr+i22sWYPJugwNRlIZGKN2/CuIpKOEsGF607c3F7gM31tW a+3lt6/W5hiSEnPBI9JJOSzS9xfWwxbupQa3sZsatQtSt0dZE/8A6B+X8OOeJlbe8h9uLb6FH898 dXQdD5nc8Z9b5hZ0K4bBkbz3kyj5cp/BjPV9Yj5P7fYPGyKC7EKFFSx4AD4ScLnPI2+hu9StZY7W ZrKORCiXaqplNfKiuCAPhYfF58Syzilg8QL0a23ZHv3UdNlkc6RDYOltLNEvIdikaoxChQe/UkCm NHbgo5l1veP3ZWnp4069/Ec7l2Ppz7WEd5YG61Bki502lxLFLI6sQDlYnugHiK4pbuODzbxaxqJ2 5Vg/6j6K10e93Vpvgba1nOjW/JmuWkHiYqKURVQA1ymoJJ4HFVdk6qjp+thM+5GOLajLGm5k9uSw t7/QL+zuIpJoZoWDxQU5rU7wCVoM1Ri0ZZXXgZWrkoSUo7UabL0+DTds6fYRxS28SRUMVxTmozMW OelRXMcROTk6hduOcnKW1jrUbC1l/TzQRvPbK/KkdQXQkGuViKj5MVKJkT/oDC+8krf3kPtxbfQo /nvjq6DofM7njPrfMJ+it1FB0/UNUu9/cZI1FXc0SgUYpql8/YS8lGt32QUX91fC7SNoEmdYXuRb M5CKsfnoP0knm8nm8+MVSgrGKptMT6zrcc5h8BHI4bJVJDQsTSik0zfJiEo8QVuHEjYbnWV1afVF s5ZDPCkZgMxaBVBFCiDsY9v3+zE5Y8Se3DiPzrutSJHl0xGpVaiTh2+etKE9nnwUjxDtQ4ghtHSN X0jc+uarHarP7RIKxlgvLEkrSFa14kHgRja5cjKKVdg3qL8bkIxeGX9ww9ta3ys7adGq8ODSZTlZ WbNQ+QBTXGNI8RPtw4nOTcOrxJIzaegjiBLsXIoAaVIPHj5MCUeJKtRe8a3e9Ll7aRvCR1yFWOdv MaHsxbtF/wAZcTvnP7OeZTjlzU+SuEd4vQrr3kPtxbfQo/nvjq6DofM7fjPrfMKehaxrskyCNDIb udS5FWy900r5sU1XWJeTf/pT0J/cRvW1WE2jrDItuCXoBwMygeQ+sRjGNKYi1qmXHiRYj3ESEzFm RlKgiIkVrRgWHdr904PgafA1gGumWO153JIjEkakKe7FQCgRWYvTgfWpiWobQahtN0TXRHlSZFSM LJVeUqBShnUcAKUHYMR8A+AznvNXglCTSlXorqMsfYeIbgONfKfLi6hF7CyhF7Dm+pX7hw8xbmAK 9VTiFBAqaeZjxxPbRbtxFJqV/JzM81TKpSVsiZmVu0FstfgwdtB20NpZKQyggZSjA8PgxZlwn/41 /wDX+Tjm7zn7yvPeQ+3Ft9Cj+e+OroOh8zt+M+t8wp6G/YX++z/k4pquv2EfJ/b7ILtU0m5urmO5 trrw0qIYzVA4Kls3YfhGME1SjE4XElRoYjburhVQaqQqUyjljhlNR9/jgrHgX7keBs+3tTaaOYX8 aSRRiJGWGpoDmzHMWq1eNcCaIV1UpQzHtvVjGy+1ahFzCsQNaDJxJ7e6fLgrHgHcjwGVxtt5bsi5 1WNroqCQ6BWydg4AgUxZXEsC8b6WxGn1Ujox9pRUWuY5eynb62J7qJ/J9Dmdt2orXVYRQgGq04kZ h634vHEd5B+T6GZdpK9vIY9QieqMQAta0H8bE91B+T6DzJ/t1y6+rlr8lMI7zCpXXvIfbi2+hR/P fHV0HQ+Z2/GfW+YU9DfsL/fZ/wAnFNV1+wj5P7fZFg4WOeLAAsAG8R7xH4ysPujEMCOvNG068mE1 xFnkAAzVI4CtB9/A0mBwO2NFP/gPZT0m/wCuIyoDeTb2lSvneIk8PWNOChf4FGDKgNRoWmWqvPDE VlVHysWJ7VIPb8eJyoBj/wAf4X3lik/ez3RqmldRbO3tBAY206JzzYlkNTJIO1vixtauyiqJmsNT ctqkXRFX6N136l6LZeC0vUILW1ztJyktYCM7UzHvKTxpiZTlJ1bMrl6U3WWLH37yXWH/ADiL/CW3 5mK1ZSpIaf176xXtld3Q3FZwG1aFeTLb2yvIJWKlkXJxEYFW+DEOTJQRDqf1ZM8Uf170dI5M4aZ4 bYKhU0UNRD6fkxGZgY0vqj1avbeG4bfej2gljWTLPFbKyllDZSAh/GI+MYMzAxe9Uuq1tY3N0u/N ImNvE0ogWC3zuVFeUg5fpnyDBmYAf+8n1h/ziL/CW35mLVZFQvs+qvVe42+urneVpGzQNM1p7PiL KwrRCwSlT58RmZIHP7yPWBo2DaxFRgQR4S27CP4mCrIqeovFT/sh8Zm/pHgufmoP5zk5607PSxkX A73jOiW596btt9c026t47aK0jtuTIshkzI7sW7oK074xZOhDRUf7sO+v7TbfyZfzcTmIyi/dh31/ abb+TL+bgzBlJbb3Qrqnt+WWXTL20jacKJc8TyBlSpC95TwqcQ5E0Ja56XdYbi0uLR7jTFt7mJoZ US0y9xqdlE4EZRQ4KhQ4w9Jer8N9f3sV3YJNqTmW7At3yM5UqWCZcq8D6uCoUM6l0n6vajYXFlc3 OnGG5QRy5bdlOUZqBaLRfTPYMFQoDH7sO+v7TbfyZfzcTmIymf3ZN+0p4uCnmpNT+DBmDKYPuwb6 Ip4m24/9sv5uIzBlPTv1a1L9mPsbL/SvC+HrQ0zcrl189K4qWLDvOzAA0wALAAsACwALAAsACwAL AAsADz+r4AP/2Q== ------=_NextPart_000_0000_01CB3ACA.1C64E450 Content-Type: image/jpeg Content-Transfer-Encoding: base64 Content-Location: http://www.stata.com/distrib/policies/webforms/mlmus2_thumb.jpg /9j/4AAQSkZJRgABAgAAZABkAAD/7AARRHVja3kAAQAEAAAAPAAA/+4ADkFkb2JlAGTAAAAAAf/b AIQABgQEBAUEBgUFBgkGBQYJCwgGBggLDAoKCwoKDBAMDAwMDAwQDA4PEA8ODBMTFBQTExwbGxsc Hx8fHx8fHx8fHwEHBwcNDA0YEBAYGhURFRofHx8fHx8fHx8fHx8fHx8fHx8fHx8fHx8fHx8fHx8f Hx8fHx8fHx8fHx8fHx8fHx8f/8AAEQgAaQBUAwERAAIRAQMRAf/EALQAAAICAwEAAAAAAAAAAAAA AAYHBAUAAgMBAQABBQEBAAAAAAAAAAAAAAAEAAECAwUGBxAAAgECBAMEBAkHBw0AAAAAAQIDEQQA EgUGITETQSIUB1FhsTJxgZHRQnMVNRfB0iMzkxYIksJTNFUnGKFSYnKCsmMkRGSERkcRAAIBAgIF CQgCAAcAAAAAAAABAhEDEgQhMUFRBWFxgcEiMkITBqGxUnKCM0MUkdHw4WLCI2MV/9oADAMBAAIR AxEAPwC5/iV8zN87V3bp9nt/UpbO2msxJLHGoYF+owrxB7BicY1ISZP8lfMjeGubSnvdZ1CS4ulv ZYldwFORY4yByHaxwJmZuMkkPF6Axv8Ae+p2yFjdkfJgZ3pLaTSqCt75lbsdj4a8ZF9NBiqeakto TCytpBbzD3x/acnqoB82BZZy5vL1ZhuMPmJven3pJ8i/NiDzl3eP5MNxofMXfFPvST5F+bEHnby8 Q/kw3HCTzI359HVZPkX5sMuIXd4/68NxDn8zvMJT3dWlHqovzYsWeubx/IhuKbXvNfzNXTZzba3N FPGpdWULyHMcvRgixnJ4lV6Cq9llhdNY2v3/ANY/A/7b8RJ9q/ZWbxvDqdbo16vKmbNx5Y2jO2F/ v001WPl+rHZ6zjgPVt2cb8aNrs7Hymxw+KcXXeD7WMt1aRsl0bejOuURq9eRrUkYO9NNzy7cm323 7kU56Kxqm4rZ9nSzvml1J2J406KU/wB7HQOCB1oOB2KDwGoPQf8ABT87FTsRJ+YzwbCWnHUH/Yp+ dhlloju6z39wR/aD/sU+fC/ViLzWczsFf7Qf9inz4jLKR5R1eZwfy/ir94Sfsk+fA8spDlJq8yq3 DtKPTbDxYumn/SJGUaNV4PwqCD2YquZeMYtrYThcbdANvrMFDwqKEfFTFEZF4zvDJ+B3hvoeD6fx dOmOoxf8deQx8Papyh5v371j+rHtOOC9Yffj8vWavDu4+cEdy2sN3s2/tJrlbOK4jeN7p60jDMgq cvHB/paVLD+d+5FeeVZdAFnZWsQxSafJuaDw8k/WuIWkZXNxUySd3NwJJzerhjp/PWugD5b3ka20 nUrW4aG33XDDOytI8UMj1NEpG9T+sUFhXjy9eJeZXwiVvlN7F9VF/dE7kS6uEhmtoDJI36OWZUHD iACGXkpr8eHerUWK3ylfcxa8tuFvN2LLF1bdJlaWUxksQYlqOFWKn1HtxNS5CLtcod22/bF7dbuS 7gltJcpSVWCn9IaLw9fZwxCj1NEnBa0yyTcWizymKK/hMiqGZS4FA1Kd493jX04HuW9ok9jKbe94 r7dd4ZElCXESkowYBg3EHKeeBL0ewy633kLqTUBkYSJwIPEYzcAUxp1X8Gq/R8NX4smOn/F9Jk+P pDbfv3rH9WPaccJ6w+/H5es1OHdx84Ibk8Gdp3S3okNqwKy9EhZADIgBUtUVDUPHBnpivk6PjfuR HO6+gW2ov5e3HibiWW9a+uuos7SKGZZMweuVFAfM/Hu/Byx1cFPkoAUiyCNM2kNL8UJrq9aGZILq eMLAtZAT1O+tFjyjmMW1lWhPDChrFFsGsR/57qNmYoQCYkAYhjlFOAj7OOF2huxyhLY7B0i70JEh 6r6fqAjuY3kkCOwGbKRwBH6w4jjdSaUKUOR8rNGQhXWdVTugrIpFD7y1APP5fRhO6xeXDYSrfy00 CNkakxEaGOhZQCGBU1AWlaHnit3mJWona/21Domy5baGc3Ec95HMGZQmUcEVKCvuqg49uAs3crF1 3FlmGGVAHu4iFbh2HGbFhg2Kf3LU/wC1/mY6X8X0mR4+kON+/esf1Y9pxwfrD78fl6zT4d3Hzgbr Wt6VY2SWuoW73ENwshdAiOhSoBDBmXF3p69GFh1enE/ch8zByl0ALJu3YTytIumFQQAtLaL3VFF+ l6MdQr6S1kFlmthwttxbEjEsEOnyuLuYTSRvErqZaZVYBnoKDEv2FvG/Xa2BJbS7PSf7PXQS+ozI J5upbRZY0rQSMc9CGqQF7TX14Hln4UxYuzq5waVtt0ppLpr3R1tFtltbodJcsPCNUQehUWQKq+rA 64nCta+wf9dgm27bWK7uTaJd+NtK+ItDGufKvPKM/f4DljRhn7VFV6HtKZ5aSdYltonmLt3WtLGo WcFyY1YxzJkUZJF5ijODQ9mKs1cjblpZfl6zWg11TX7LVtOfT7aGZHLpNmlVVUCPieTManGfms1D BoYXG1JNNgTqEdA3qGBoMsaGb/8AGP8Axv5mOo/F9Jj+PpDbfv3rH9WPaccJ6w+/H5es0+Hdx84q PMskWEFOGaN0r/rOBiHA/tr5n1BT74tfCKq8jwx0OIsqd9v20EuvWscjQqhYlxcGkbKBVl5ipIxC /Nxg2q9BVeeim8Y+zIRNa32o0Wt9dzBCpLAQ27GGNVY8coysy/DjHzk6SUPhiv5ekFsqqb3sIfBV 4UrXAqmWsEN8bau1Rde0tKajYisqKP1sI94U7Svswfk7y7ku6/YyLB7y5S2ubjct1aR9OxnktpRH Tgs7oxkAwdxCTjbgpa9P8V0FeUjScqagls4KaklBQUavwUxm3Jdg0J6gU3fqMWmWF5dvT9GGEY9L ngo+XGhlLbnJRA71xQi2MOtx/h8zUPiPs3NTtz9Gvtx1uinIYunpGLv371j+rHtOPPvWH34/L1mx w7uPnFZ5iRl7WzHZ3if5YxVwR0tdL6gt98BmtajlX4cbakSNdJc2e4tNuGkihjMpjkmnAKIHQjMK lRm7Bhr6xWpKjejYV3NVRgbIljhl1LRHZjJa3El3as6GIy2t05fOqsBwSbOh9HD04x89V4bmxpJ7 aNf5UYPZ0Vj/AIowzigFPZgSEibRu1oTzHHs4YJSIATqWgXO3obl9EtUksriVrm5sgMsgdqZmibk Rw9w8uzBUriuNeY6NaE9nT/ZdYwoiaZdwvpsmryAwxFGI6wyFQpIOavLiMQuxeNW1pLbkhU393+9 WsZwQuhWRZos3A3U3EZlB45V9OOhhD9eH/ZL2Iy2/On/AKI+0e+X+5bLU/1Wle33MbP4vpBPH0hx v371j+rHtOOD9Yffj8vWafDu4+cW+9Yupa2w9CuafAwOBuEOlv6n1BT7wKR2xI4LUnkoxrOZYaXe 33u7Z42A6nBo2pwV1OZTXnzGFHMKLGcalxYre7igtb61mMe97F26kMbVjs41fJIzwVq1vPwDKTmb gQeGKZwVqqpWxL+Xu0/Etj1ICnF119sv9K8ytFVRHuBfsa5VmjNw+aTT5HRijdK7Ay0zKeEmUjAk uGzTrb7a3eLpj/QvPXi0e7+Qlg3bs+QA/bmnBSK5jdwUp6a58EZexcbo4yXQyM5x11RQbg8yNlvD LaaHcy7j1ROIs9Hhe6AA5l51AhRf9ItjRucKnOD1R5yiOZipb+YU2u3O4N0QS/aRjs9MiZunoVtJ maSRD7t3OtP5K8MTy8LWXfZrKb8b/wBqLJRncVXoXw/2yDdSac9pAk8CW8lso6dsKViNKdzLzHrG LViTdHWu3eTpFpJqjWzcObh+CXvGnhPe7f1fPG/+L6TN8fSHW/fvWP6se044P1h9+Py9ZqcO7j5w K1y069vbt/m5x6e0YzuHzpaXO+oLl3ijtNLAkIpxrw+D0YPuXtBbEuYNJBogFfT6zgGeZ2k6Gmo7 Ivml+1tAujpe4Y4nhjvFAIdJFylXBB7D3WpwxblOKODpJYre4pvWVJGXG8NCtzpWm7v29Pp89ohM s0XdsZp+MSd6rKyZGL1b6WN9vL3beKC7W3D3vaZvl3E9egD7jXPLV9E1K8sdOtmvhcS3cV5c29vW GNZwUjymPvLlXL8eFBXYXY0cqKmiuh6OcZwi4Nun8HLdXmILq4guEnS10981qbDT06VuQ3fTOqCr 5SOBPpxfKd6+mp8+4lGELbTXMB1zqri7A06AN4yrTgCjCXkGyj3iw4HEo2+z2nq1FmlS0LQ9Ze6Z 5ba/rk0N5e2409I1OSecEPQjiVj94/HTDxuU7KehkJyinXXJDY+zo/ws+zs56eTw/U7aUy5sb/4+ gzcXaryhZv371j+rHtOOC9Yffj8vWavDu4+cGblUNtFmqOLj1caYx8r9pc76gzxEe3snWXPl4H6R /Ji2cqqhOMqBDY2SqAQOJwFRtk5SoXFtZ1IBGCbdmoJcukqbToZYjHLGssbe8jqGU/EeGDPJppQP 5out5eXm2AxnXTIUzcW6amPn6lIGFDOXrc8OJ0YdZUZxAh9q7chevglanMMzkf5TjShm7j1slKzE MNs32gRJ0La0gs5hwZURVJPpzczg6Eqmfdg0X9y2ZGPvKQePbiygMRf/AED/AG/y46L8X0gfiCLf v3rH9WPaccJ6w+/H5es1uHdx84JX+pxWEdtI8Et0SzDw8IzPlJ4sF7eWM3h9rHbSqlpevoCLro6k ldyBkhaPQdUbq0AHQoEqad/iaDB3/n18cP5Kv2KbGdo92XFvdPC2hagYoiQZljqpoWHd4Cvuj5cR XD0lXHEZ329jC/QdRGpaal54SeyLMy+Hu06coymlStTwPMYIhl1GVKqXMCSm2WVCMHTsIqUiq3BZ rcWcgp3gK/kxh563ortRo5O5SVBOa0DDK3qNCMF5Z4kaUgdvLpo2DoxVl4gjGjbQPJE/TvMO7tEy T/p4aUPY4wVF7wK5YWwPvGx/hh46h6WXrU7aUzY6L8X0mVTtU5QU/iX8xd07Y3fYWmkSwxwTWYlk EsKSnNnYcC3wYpzPCstmXiuxxNcrFHMThoi6Cam86d+zTRzSXFqZYQRGwtoxQE1PLEYcFykY4VDs 87H/AHbu8sYf4i/NeBAkWoWyqOAHhIT7Rh48Hyq1Q9rIPM3HrZc6H54ebesi4P2/ptn4fJ/WbWBM +fN7nDjTLxwpcIyr1w94lmJ7yzu/NbzatLJ7u33hoV3EvExwxQ5zSQRmiOoPCpb4Bh4cJy0dUPax SvTetnWfzV84ItPubwbw0KUW6sywpDb9STLmplBA94KDx7Di39Cz8PvI+ZLebyeZXm2+RH3noDB4 jISIoMgHHuFqe9wpTFM+EZaWuHtZKN+a1MEdzb33nawSXM2s6TqDdboyRQQJ1K1cZwo5p+j971jC t8Iy0FSMKdLLnxC98XuK/bG4ty7k1B7E3en2RSIyiWe3GVqEDKMp5muLv/PsrwjfvXn4iHujW9Z0 TVJNPaawvXRFd5obfKoZgarQnmMTWStfCM85d3j98TJ+AXiqL1Ps/qUp3a9KvL0YswqlNhXievae fxFeUO5t4bis9W0ue3S3t7UQNFL1OoXDs1RlVlpRsTToQaqKD/Dpvz+ktvkl/Mw+IbCZ/h035/SW 3yS/mYWIWE72P8P+/wCzvbe8jazeW2kWWNZUldCyHMMy5eIqOWFiFhCufZXnBMioYtGTKFCtHaOp 7hUqeVPoD4e3DVHoRLfy3817fxBij0kG4ZHdjbyMQY0RBkqO7wjUnCqLCdoNg+bcMaxrDozqrZxn tZGObu86jiO4DlPd9WFUWEHdW8hvMDU9Rmv5RYwSz0LxwJMkYIUL3VKmnLD4hsJFH8O2/VNVmtwf V1h7FwsQsJjfw67+bnNbknhU9Y/zcLELCPn9ytZ/B793KL47wnhc9GyZunkzcq0riJIbt7yww5Dw hGYQjMIRmEIzCEZhCMwhGYQjMIRJ/wCnwhH/2Q== ------=_NextPart_000_0000_01CB3ACA.1C64E450 Content-Type: image/gif Content-Transfer-Encoding: base64 Content-Location: http://www.stata.com/distrib/policies/webforms/regmodcdvs_thumb.gif R0lGODlhVABpAMQAAG2rXw8ZIK6vrYiRedXosujyzJ+/jP39/WmvP2tyYq3ScJvKT63RWdPjmYa+ ZL/ZctnpzsXdiPb67tTU1MLbserr6e713d/tvY7DRoG8PlOcQjmGMkJHSVZhT3e2RScvNSH5BAAA AAAALAAAAABUAGkAAAX/4CGOZGmeaKqubDtKByzH9GzX+K3n/D6zg0RncCE0IgpkZIk0OJ8Gx8Dh TDKXCgX0ab1GjteGGJtdigmEiwR2knQ+Hw4h8lAs7gsFY+FweACAGoIagA55WXt8fgAeHn6GeQyS DA+Vk5WVdn2HD0wNFRUoFRwBAR8NmQsYGawYrhkeCAiDgx4ZeK6Ns7uyshkYeHeTe5d7rLauDAqd ERMVbCWjpRx0Y3bXdncYsbu2eXV23IK8sd6SmJTozJK435VLE84n0gEcy2kXDV1IuI3Id1ccaJg1 qFe5W5TEoHqgkKGYOlgWdDqyLIIAeSbofXCgQEw+hkpUTdIWrFMWgbRk/+lqhNAhHZcO9WU52YWS gouhMnL4UK/PFzQRjCxRtUCZnkQYOGaJQmsgAn+NgDH48pAqKoXKkCjAUDMJznlRqAAqc0RoBAfA lllS1UiLE5TjnkL9VRTTg3R2JSV19KjopJsYozmQ5QAAIQCG1OopWjHVql67UkL9h/fSlwiJYEF+ Stfo14yDERQumAGvnqAN5oBEAjelSn+t6t6dpLDBnlXcDNJl/FlwrMIbCjqg5MkIRTARxI0bZwtW bNmWGDS8nVu3K96BSVQIDbygB4mX0VCUDmZw09esWkmtnO6lqgybrWPPKZgwgOAFuVIVz49qBAyy DBRXc+kB4xdes7GDh/8f8XGWVm/acXcfLR7o15AdySCyTYBOEajegXmNhMFUJwkY11MPZjfCdvZt gB8Ct1SkED+xvcKhU+l9CGJ0DLxyiAIeNEVaHhCuKKGLA1XIWCcEFHBBPmUwsRQVVcyUBTNW1bYQ RARQQEFYfYRZRQQUFCkCi6INgJ+SwtxjAjQlwMnCGnLKGeeKZh6ApkDBwajNdR3FYIEFTxZaKBqI ImpoAYw26ugFjUKQqJeUVgoBpBDkuaeaKLKFzFQHWMBoEYlOSmmihi7qZBFOFiBppbCemkamKp4p oQZKbrOJHQ+EOiqqh3ZZaqqjNgoppK5CQEGXlRoAaxquasqdH8Bs44H/AVYsEIGghJKKRhFePonP k44WAG0BohbQgJcQuNrlBbGeSi6t9EVImAK3bAgAtklwxK24ihLAQQcGDFCwEwIYkPDCCjdcwAAd DGFwEE4kMMAAlRJwKbr0zsMdhvANNKa/g4oKcBoCd5AAByurvLIQEAsRMcsWDBCHzEKwLDMFkrY7 KMfS3vvYIPwqwJEEJa9a6sUUs6xwAgJYbHEQTBcQtdRUTx3EBOYyWrIFHYMm9B9NAYIAABK0Oqq3 48KLcbHlkpsushBIamy7rkL6tQUVBC2a0Vr0kRIhEkiqd7pNer0qo17+7KSzTlJg7AXtwiv5qsuS i27JYdcnWh9PtCYI/9qGby5q4uiuCqnEAgxAgAAcFAzxxQYkIDvTE1NcBLJed27v51QKjt/oYC9r utqpF1qAzDQbQHDECbzMMgcDQ+yy08o76ruRhBUWupAAFJ/G17ybjKwAraNfAAVYM+381E7gfjEF qRa6va32HRbI8ISIz/vkcEud4wD4JApgjFmGK1XbEFWmWunpVoPg3+gqoKzMHatg3/pfASQgAGQV wQAEeMK3nOUlZ4EQWGzrkt8EJyRahK+C+HBSxLImv9lJTQgEGADLLhY9qrmsh647mZdU6MA9GaaF gwAABYcIrQuoDHoDY1nEoPdElYXwAzp0WQcGRkWVcUBypILVCkU3OP/SMXFz6CsTxMqkMPShbwAJ c4KXJgYFOsYPg2EUYxGndZ7NeKB4y2qSqJJmN6/tbW0ZLIKkxnWuX6EhVmN0Snx+U7hANhJumitX AQ5WwiFSCn5dIgLKYkXEenHvc36cDAYKx0jeHepw5arZE4MAs9hRgAM27GEHAhmvMb5mMlGRAKuc FMO1yS1djjIALqG3Mpu57oc62+KySNlAU+IPlcBkSYVY+cr/ZU+T5jKhCb9EsBCK00sJm2Yv99i9 bLLkF8Jk5CUXB87JKU9YinIluOJVTY91Ty7AbIUE0ODIb2byWI9imxAT6S1+9lNsn9OFQd65ykEi 03Q/GxTS9vY1cNb/DW8D/KhIPzoBX06yORiwAJ1WytIYsPSlMEWaTF+qUo5yzqSQgY16pkIHSSTh SQ0A6kuY8Q7/5AUPykAHLvwwmT7gNKceusUSHuCXoBgBqHbJKpaio4rrrMUVr5CoQRzw1F/mKC0J SohCjOMFTJjhCGsRiVJX8QpYTCYDZGUnKlPpIWBkYhjMUIi4kEOHyxhgqtm4gxaMRi2w2hUqSSkr QHMkVTrsyCrHqo0Z3Ja16FnsdkAcAL50+ou8WvOB7QwTU9+ZnkMkwjJUKZSWUpMPONYQW0zzgPu2 Qlm6jJEKi50LgUaEIQOdYz9ATY1xUgXX9mBmAaywEr4o6wrT+hN4/4Bb7VzwmoUFIMO4zECWcvM5 2KFOtUewmEkdqJui0+6pD4sFkju5YjS5xIY4+DiDuIqgpbdaFjeQuMNZr7OAMXogu+5sRB8MoxK0 yCZLiTTObI+AXlgkBRIFMhCR9NoHgLJ2tU+JwhE1oBQHgOQdaz1oPs5gBKquIrpWejGBN+zeaXn4 wDNhaogNMOJ9RSEJdwmsfienXyOADMYzAauGaXzdDue0u1vpxUAM46LgAIAwkXDuGR5F2yIg4cWs +BFvwdqmMSrAkxQ4c926xGMvEeJLGiChlyKwgEaKIaNIMyTSCMAQ8VDgC8cygrbmMJ8mO4AAhYOA ShVQUwgUDgAUOP9AYUChAUfzrAIFWMAa0HWAfMAAUjG4gAjWcIACbMtVBTjAmTut6ANIp9SFhmgf NJZeCyzAAhSAhcY0sIEKGGADSqz0r+Nsawtg5gGimkraLmCBUiPtUmmLgASgu8qgSEAPD5BAj5oU a88ZgtYZYDQGTC0LClQAEBb4dbCV5SLJafrYhFK2k5qdalE5Wl3avoUEgiqBwl67zgXw6RjTEF0L pDQC5d6OAyRwn3UP2wDFhvcFKBFPevtKDfjeiqb5XVhXLwBS3f7dtyFQ8HFToNwWQIyvgV0BDUyA AsFxd78lQag8VLzUvro3oqmq6TT0ewn/BrkyVgjwkpMbAebuw8r/HS4ICkSc5hewuRFSXW9om0sC VGXAQNXAhKBnmsmyLnq4Db6+WZgbEEtveQUoMJAIZNrYULc5ZtSA6G7teeYK6DeiT6NqgCuW6GlY BaMXcHSeCWLlBji3slB+66zXPO+o2CDG0YA0NSxk6z/f1h5ADnZv15nk4b62BQjgh6Q7wNf7kkDi T54Btze+KDXXeuSFKczUVH7fnRDmBsHQd6EDpsbdA3fekZ1mSCtc9amHeK4x0IC324EBo7e5Puhe ewIMStRg0P2+j4D1jwd86Bw+NARswWhGkx7SFlA6jw2gelz/ogHF1nj0Za8Ay9e+AZU/gHG0r5AD 3IXZAhd+adAW/1iHawpgfH2AfBQgegTgCtxmAfJHAHJXZ/02UF9wffonBvz3Bf7HAMz2DoA3fggw eLgWC+ZGCL5GYhJwehK4AM13axE4ItfGfFg3UBpIKJ2GBgfAb5XQADv4AOgCggKYaRgQAWSHcAiw axqQguxnfnfwghB4C6Mng/qAaPhnLmogKr6Sak3SbwzRafBnAUIIfJ/jdEbjdCYnC0qYbg6wgEbY glCIL7cmgQywg/W3b2njc1qYLp22QWYAhoMyhoZmek7nARCAhO0iC+nmAW44UKoQh1IYddnGZ/a3 Knm2hX34cz6YD4NiEysEaRAAOikHAQbQdBDwFMbWhhUIh843Iv/x1itzsHvLloX1xmmQsm8caASd CH5kWBg8AwiJJ2wxBwEDUQF0Jm3SJoHMV2yOVwC5hwpcd38Y2CqdplJl8YOM4hkcBoqCcHrCuAGG t4QR0Hr9VnMuaAHQB28BJ21VWIELcToZSACZ2H9z932/12SgCIwtR4rtBgHBcYgKgGiINnEMYH15 p44UR4nuWAmMgn9GMJC7FxQHUI/a2Iv5uC/7+Gvg6I9LmGbyqAYXcBfWl22YAX0TN4kPEI2owAAN aYOb6IdfGHnoyIv4eHYY+Y3haIx3CJKVMJLGdgevOFApuZAs6WkKMVAb1IPYKIZesY084z0ZOYwD YWwRIGoY15P/6Ih1wOAkCTmU+GcbRcmOYqB/anAX20IHQVgHnxiO3kiKpoiKqGCFwiSS6CiGW3mS M7dz+yYJTsKOX4B5Zul/QGhsnbCWxLgBp3d6FFB66cdotoF/RhiSBVmX+DJukoh7friXHxeSuPiV ZQkqd4EuTGCY3Zhuo9cHZoiS+IdsBNkk0FeZ5iIRNiiLdPZxfOaFnol1yvZ/xrYupKmCyld61zaJ 0sGARWF9r8kKsTmJzYeHtVkEw5cE+6YG56BqHtibFhF+ZugsoldBjiadBKBs0GdqzZeVafaGEgGI FXgBSzB6efeQO4h9qWGdH0gmZlYBnNOGmIYuBLAM6kIJEFBn/+RznSVDZ8h2Vc7IbAXKkiXTb/iX NHqQBswgLazAERWqC1fmCBp2DgriF+hwG+BFVGmVVs6lVXZhBpoiM1U0RSzKokLwoi0aozHqWdHz RDCqoio6RTS6o3BURDxRCkAapEI6pEQKB0b6o3BgCkS6pNOwRdTDpBygKaTApFQqpFvUAa1Do1L0 Rj9apaXwARfhRlM6pFFaRGPqpVS6RR+QMzvRpgMTPV3qpWvKNAkQp0Fapu51pmi6pEd6pKbgp3ZK pR8APRATqNMgpXsqqFHTOnAkAFsER1iTAIn6p0bKp3g6D3o6qUBqozQKBz80BB2gqWh6qTohqqZ6 qntKqtGQqYeo2qqtqqrawaquOquTGgeISqu4aqqwuiJv0Ke++qvAGqzCOqzECgdQU0Tx4Ea2lT7L 2qzM+qzOGq3QOq3SKq3oEw+nJQEVkKxu1K3e+q3gGq7iOq7k6q3YaicxAArxsK7s2q7u+q7wGq/y Oq/uCgoqoK2gkK/6uq/82q/++q8AG7D7SmoiEAIAOw== ------=_NextPart_000_0000_01CB3ACA.1C64E450 Content-Type: image/jpeg Content-Transfer-Encoding: base64 Content-Location: http://www.stata.com/distrib/policies/webforms/tips2_thumb.jpg /9j/4AAQSkZJRgABAgAAZABkAAD/7AARRHVja3kAAQAEAAAAPAAA/+4ADkFkb2JlAGTAAAAAAf/b AIQABgQEBAUEBgUFBgkGBQYJCwgGBggLDAoKCwoKDBAMDAwMDAwQDA4PEA8ODBMTFBQTExwbGxsc Hx8fHx8fHx8fHwEHBwcNDA0YEBAYGhURFRofHx8fHx8fHx8fHx8fHx8fHx8fHx8fHx8fHx8fHx8f Hx8fHx8fHx8fHx8fHx8fHx8f/8AAEQgAaQBUAwERAAIRAQMRAf/EALQAAAIDAQEBAQAAAAAAAAAA AAQFAAMGAgcBCAEAAgMBAQAAAAAAAAAAAAAAAAMCBAUBBhAAAgECAwQECQcHCQkAAAAAAQIDEQQA EgUhMRMGQVEiFGFx0TKydBUHCIGRQlIWNhexIzNjg8Mk8KFykzREVGQlwWKCokNTJicYEQACAQID BAgFAgUFAAAAAAAAAQIRAyESBDFBUXFhkcEiMhMUBYGhMyQVYiPwsdFCUuGCQ0RF/9oADAMBAAIR AxEAPwDn4qOZ+ftL59soOXr7ULa0awjaRLMyiMvxJKk5NlaUxOKISlTeIeSedOfm5Whk1LVNQ7y1 xOCZpJFcoMlPO2034sWYJ1MzXX3Fqj3Bk/N3OTNSPXL1CdwMr+XD/KjwMz1NyuEmUxc1c9kVfW70 U3/nX8uBWUdesnxYbFzbzpQf6xdt+1fy4l5S4C/Vz/yZ1ec188JHnTVrzZv/ADr+XHHbjwJy1U0v EwYc38/MoprF3t/Wtjis9Ap6y5TxMti5s56Zwvti8Ndn6VvLiStLgd9ZOtMzHEWuc75B/ql2fDxH 8uFSjHgX4XJ02s4k5i53jcH2pdFekcRvLiORcDrvTW8132x178LK95l79xeH3rMeL/aK+dv3bPFi vTv0NLM/Jr0CD4kHcc62tGI/gk3Gn03xoaHwPmYnvX1VyMjoWm3l/pVtwS2x5xUnZWqnecWHCsmZ Vy/kiktrDI9GgiOW7uMjjpTtsD4lrjmVLaJ827J4RoPNLk0eLKl1azXMZ82dY+HWm8HMQDiEqbi3 Yhcb7+zoGitytE+aLTGcPtGaROg0+iGphLuSL60cFjUo1TXdLy93i0pSGXaQxqOsebicE6VE6iMf ChOuoaLHlU2JqPrOR0/0cNU3wRnemtrbKRdb6jp5uBLDYjf/ANzZ82XEczG29LbcqqUhuNetCCps Vr1CQj5qLhLqacbUdlWdrqNvMMi6WCTuCszN81Py4W22PjaiuIX3ZPspwOH/AHv9HX6WetK/0sVf 7zVw8nooJPiR++tr6knpvjQ0PgfMw/evqrkZ/la4mHL0VuFEkLTTu0bVoKZasSB1YsuFXUz1cpFK ie1l1y0ZdFgjCyV372+QYl5TpiK9Sq91Gr07lXVL2wWR0Ef0qytSlfAanCHgzTjHPCrZxPy7a20a vNqCxgnbw0L1pv3kYarbZSu6i3bW0baPovL8kZn7xLNViuYZU2KPEThV6DWBb0mphNZliB6jy7y4 1wsn5yQk0yu5PT4KYlCiRW1MXKVS+GDlyFQwsoyPqkE7vGTjkqDLbaVdhdDc6W8iyWtrCoG/sCoP VswqUabR0JObrEuOtuqsiQsh3EqlBiDVRscNoBxX+zPEoc3fq06a8TFSn7hsV/Y+Aj+JH762vqSe m+L2h8D5mL719VchJyfa3Fxy/DHDA87yXEqAIK0qUJqfkxftUq6mPecsqSTdTeW3K6xzLw04dAOJ ckUZj1Cu0CnRhkrsVHpFrRylPDBcRsbkBDZ2TKTF2ZMzbSaYqNN96RqRpTJBrAyur6fc50tXdXlj jZnymtBmrQePFq3Rqpk6m001DCqQ406yMHDgACRFSYYwDmJp2mauKlzE1tNHJSOxUwW/pZTdMiTB aVCntU34lG1VELupSlRFsWmRPbgwGtCTQ+HbhcsHiMhFOPdCtPs42H6IIR51K4XNFiw1wDDZOfNV anzR48QoPq9wr7vL7I4Gzi9/p4K564pf8hp0fk9NDM/Ej99bX1JPTfF/Q+B8zE96+quRf7pbyW05 akmRS38RIgI20LAf7Bi5G2ptplBX3bgmjb3etWcoZpy0jrCcsCK6xg0QVzhQwc9rs1bbU1ocI8qc XhxLy1FucVm25f4+PQDDUNAtY3S2iZI88SuwEmYB8rS0oGPZzMpPgriTjek8cRcZ6WEWo4bOeO0W rf6A0U8qxqklvDJLWsgd5W4PDiJc9sM7Mo6dmGyheTVd+H8yrCelaeXcm3jjuwx212DC61ezW3K5 TDcukqLmVi7ExnIC6DKozU8tcLdmdegsLVWqYujxWPLYJtbutNbWJnsnYRKF4ClWBIy9ovnoa5v5 sXNNbnk7yxMX3O/aV79t7i/Sr+YMeHDmalcldg/l1YjesriT0esluQVd6xrwjZu55Vy1UBN7dWau Kzsx4l5629vj8g/QNSmmMffY+C9KgjbU+Lowu5aosCzpNU5PvKgPnj4Wep4ftCtabd/VjLp+5Q9H mXk16DG/Ej99bX1JPTfF/Q+B8zD96+quQt5K1Ce15TRYpcma7lZlAqSAEFdvjxqaW3mbwPP6++4Q ik6bewfvr0scyrL2o6Dx0OzbT58Wnp18SgtfJPHZ8xnZpawoZ4S8gkNVRWGVQRtG0HpwqbcsGW7M YQWaNXUWQSaZca1bPdBoLKZ1ExXtui5suYUG/wAQw6eaNt0xkihb8u5fi33YS2mslt4St7cypwIo rEz25t5jdW1wRMI1dCe12QwDA7c2MhXpJJbXm4Ua6D1ctLCUpPFLI2sc0ZY7fhwA5+WbGOaW4u7i QLJa3ksZ4TK4ktVUmqk1y9uorQnwYd650pFYpr5lJ+yRz5pywlGW6mK7Cm4tI9J061nV2YXL8IZ4 2jIfhiTN2jVl7VM1BU+DE4XvNm402C7ul9NZU06p9FAmxnuBFSdi4Y0ILKwB6tmI3ILcN01yVO9i XrIA3EVaSJtp1jd0YhlHZqY0xQB3key+8ZTTv+bL9Lz92MnL+9TpPR+Z9rm6DL/Ej99bX1JPTfFr Q+B8zK96+quRnOWS/wBn4VWpHGmJ3da42tG6J8zyvuaxjy7RhdSUysT2jsIr4MW2zMpUb2V+ILPa CHIy5MIkszLlq5kg+IKL2W2u7e6tZGiuLdg8Ui0NGBqDt2YnKKkmnsZXhcduSlHCUdg9s9e1GXiG S4RWeKS3EccMSRJGzCVssYXKC0naLb64oz0kVx3bzese6XJf4raqJUXH+YLcc06m12H7whJ4+ZRB EEPeQBOWXLRjJl2lsTjooJUx3b+Gwr3febrkmqYV3ccH1hMFxf67HKt1f28fdFe8maVVhzZIgjys 0aEuwjUVriErcLDzJPHDiNheva2Lg3FZe892xberadWttqVosgS3mniATi3KoxQMyqy0zAH6a9GO yvQnvp0ELWmu2apJyWFWsUEFtVgJW7Jjm85oypVsp+fEFKL2bB7hej48H1FfH/8AGON/na1/aYyK fcU6T1Gb7Gv6e0R/Ej99bX1JPTfDtD4HzKPvX1VyMjosxTQ7cV2cSdqDrquNbSvaeZ9wjXLyLBdM xq4FB14sZyj5dNgZFeSOwB2rTfuxJSFuITHcI1aEZuivXjjZ2MTl9QmikLo2XrwOjRJSadUDyXBd 8xbtHbXdgRBxbeIZpusvapdoFDm7tZrQliRlE65S2zfl34VdhnSXBp9Ra0l/yZN0rWLj1mhfnONr U5Lal9IXQyNIypHEYreI0y0ZncW3gy4oy0bzbe7/AKv+pt2/d4q3s7+z5JdgLqHMMl7qDXYijt43 VUKI2ZmKkkvI2VAWOboUfKanDrVnJGm0q6rXebNSSogzjj7Dcauzvda/tcZFPufieozfYV/T2i/4 kfvra+pJ6b4bofA+ZU96+quRi9Lami2w/WTflXGnYe087rVXKfHnANSd20gYe5FRQIl6KbTQ4FMH aLUnG/HVIg4EluDSgPjwSYRgV943YipEsh3HcEHfXw46mccAjvROw7a4KkaEF02YBflxxsEjV1b8 MK17Xef32MKv3XxPb0f42n6e05+JH762vqSem+GaHwPmJ96+quRgIZQmiWu2lXm3eNcaFt0qYeoj XKD8Qu429mlKHEqictEdBF+saY7QjU+tKV2BjXrxKoKNSyObMNpqRjqkRcKFhcYlUhQ+NJsFMcbB RIk5O/HEzrgXJMN/Tgqcymz4n/qbP/mP3+MT/s/HsPZf+f8A7e0s+JH762vqSem+G6HwPmVvevqr kec5yNKsx/vTH/mGLsHizHvLBBwtOWXdFbV5I1ZFMn8K7FWNMy1DCoHXiLnLgNVi1tz/ACOo7Tls 265tUlilbendmcAdO0EYM8+AKzZarnx5CeUKLhhFJmjDEIxFCyg0Bp0VHRhqKzSVVtLD2SGr48dq La3FnFGzqO7EsxDKdK1SDgzAonDbCSPlGOHTqOXw4MxzKbvMPwer+v8A3+Mev3HxPW0+wp+ntFfx X813Okc/2VvFaW9wr6fG5ebi5gTI4p2JEFNnVhNm9KKoi/qdFbuyrJY8zxd/eZqbRRxHTrLJFmyA Cf6Zqf8Aq4atVNFd+2WWtj6xjy/zFzHr109vYafp3FjCluK06DtGg28U9WB6u5xBe02OD6xpcvzl b21xcvZ6QY7VWabLPKSMjBSB+dodp69uD1lwPxFjg+sF1DUeabE3PFstJkeyiM06RzSuwUSGPYOK CWqp2Do27sHrLgfibHB9YbG/PM9q8/s7SkSOMSyI0syuqkVoV4u/rGD1lw5+IscH1kD86cQxpaaK 4S3W6DpcuymJpGiBWk1ScyGo30ocHrbhx+z6fg+sJji58e7NqlpohmXLUd5kC0kClTn4uWhzDpwe suB+H0/B9Yt5c1rmXmC4vYLeHR7WWy2yrdSXEebtFTwwrvmoRtwesuB+G0/B9Yt1vnTWdH1SbT5b XTLiSAjNNbm4eJqgHssZF69uzB6u4H4ixwfWeq+3rj/529scGLj5+Jwe3wq96y087NT/AIsV/Mef NvL3poeV5f8AaOvif90/OvNXOVpq+j28MlhFZR27M8oR+IruxGSm6jDbhalQe1U8Y/AL3lf4OH+u HkxLOcyk/AL3lf4OH+uHkwZgymk0D3Ze8HSNLaxblbSr2RnZzeXLZ5gGKGincMvDGXZ0nrxHMdoE w+773hR6XBY/ZHRWeGtbs/pZKuWpIekCuXxAYMwUPq+7/n8SiQ8oaOWyhCM2yivn2ADYeiv1dmDM FBdp/um5/smJTl2wkBijiIklzdpGLFwSDtau35OrBmCgv1j3J+8fUr5rpdIs7NWSNO728gWMcNAm YCm9suZvDjuYMpTae5D3r2bl7SJbdzvaK4ynZ4QMGYMpXP7ifejcStNPbRyyvteR56sT4SRjuY5l PXvw65q/AP7L93X2vWmTN2P7RxfPp9XEK4kqH6R1nzRjh0T4AJgAmACYAJgAmACYAJgAmAAz+4/L gA//2Q== ------=_NextPart_000_0000_01CB3ACA.1C64E450 Content-Type: image/jpeg Content-Transfer-Encoding: base64 Content-Location: http://www.stata.com/distrib/policies/webforms/splp_thumb.jpg /9j/4AAQSkZJRgABAgAAZABkAAD/7AARRHVja3kAAQAEAAAAPAAA/+4ADkFkb2JlAGTAAAAAAf/b AIQABgQEBAUEBgUFBgkGBQYJCwgGBggLDAoKCwoKDBAMDAwMDAwQDA4PEA8ODBMTFBQTExwbGxsc Hx8fHx8fHx8fHwEHBwcNDA0YEBAYGhURFRofHx8fHx8fHx8fHx8fHx8fHx8fHx8fHx8fHx8fHx8f Hx8fHx8fHx8fHx8fHx8fHx8f/8AAEQgAaQBUAwERAAIRAQMRAf/EAKkAAAAHAQEAAAAAAAAAAAAA AAABAgMEBQYIBwEAAgMBAQAAAAAAAAAAAAAAAAIBAwQFBhAAAgECBAMDBQoICwkAAAAAAQIDEQQA IRIFMRMGQSIHUZEUFghhMkJistIjdBU2cYGhohcnGCjBcjNDU4PDNDVVZeGCc5Oj00RUZBEAAgEC AwYHAAMBAAAAAAAAAAECEQMhEgUxUXGBEwRBYZEiMkIzI0MUFf/aAAwDAQACEQMRAD8An+0/4oeI HSfWG3WXTe5TWdrNZCWWONAwL8xhXMHsGGjQV1PI7v2g/GKHbbCU7/Oksss4kJRQSqFNPFfjHA6E kVfaJ8XyhcdSTmTWV5WmP3uknVw7DhlFUr41Fq60EP7RPjKiEnqSY1ClTpTga/FxDjQlOo4PaI8Z hBq9YJiC0feolRqBypprnhlH2+hFcREPtF+MLV5nUk6qGoTpTLI/FwkUvEZsei9orxj0ktv83AEd 1fJXyduLIxVKiN4i7f2g/GTX9N1DMUdS6FQhAFSOwfFw0Le/xIlPcRz7RnjHHn6wzSDUy00rnSnb TFVKD1qCP2jfGNpmr1DLpDDuaV4FqUyGBLEHsOof0gb1+gr7d9If7X+ydfpmXM5/IrzOFK6s8IMY z2l7q5i6usFileMGzBIViB79vJjvaVFODqvE4GrSamqPwOe+vkvLuHZ2LPIV9IGompHfTy4zapD+ RU3GvS5Vtuu8zx269FgByFKNMfp6jXXSciP9uMmV9PZ47TbVZ9vhsIj2V40ZUrQqqqFzzArilptF iaHU2689DEpjcsJIwi1OkgA8fdwyTy+hDeIVnY3auSqAvX3jVp70njgtp1CbVB+322aSNk96SAQB n74YeKwoLJ0JS7K6KGGtWEbczVUita93h21xbkSS4FedtsgWtjcnmVpHI9eWzZZ1BOf4MURTxLpM L0G4a9kKghDIMzXsYHENe4K4HWug/s56O37Lp/0cVeJYQPac++G3/Ux8tsd/SPg+J57V/wBFwPM4 tkG47RaSUqY5Z18/LOI76FZrgW6fKlt8QDpJRDQBg4cnT8HhxpwrjL08DZnxEydJayrMtTQVJwrt 1JUySvSrC1aHSOUzBqEZ1FMWK28lPARz91SL6od0qE4niPcxV0izqBHo9tQIFCtKEYOkHUJMvTUs oQNRgilSB5qH3csWzg2lUSMkqjHqfHkGSiVJ4YrVmozuheqRQv3MuPmwK0HUPb/Rh+hL0bs9E5f5 lMc+nup5myvtM/7Tn3w2/wCpj5bY7ukfB8Tg6v8AouBT9A7f6R0o8wQMY7qRQDkM1j8n4cW9z8+Q vaSpDmap9gijXv0RNQFWyFSaDz1xQjRmD9XU7EwYBmENsA9IRdHwGPmIwYBmBL06gjeqUyNaVB84 wKgZghsiekSRtpBqBGtACe4GanlwUROYVFsAOvu8HYeY4GRmFN0+qqSFzAJGAnMIn6eURSto4Kx/ IcGBGYv+V+rDlfF0/kpjjf2czrfTkZD2nPvht/1MfLbHb0j4PicPV/0XAHhKNXR1xH3RqvZM2r2J Eez8GLe5+fIr7X4czbSTpOXi0qvLeNtRDUOYfIke5TGehoHeenkT87EUAZmVJZI2DKminAsK0dWz IpkdNDiQDW458BJRYi2oaW1VFCR/BXBQBsMrX0kn0f0ZBpnnqj0/wYPAB6N1TVkh1Oz/AAvhGtMA BT3CiCVqIaIxpU9grgoSJuZ6wEKisZO41Ce6GFC2dMlwAHT9XtMvfcezjjj/ANnM6/05GJ9pz74b f9THy2x29I+D4nD1f9FwKbw/3L0TpspWmq6lan4EjGNN9e/kJ2vw5l3P1jtlvM0M97FHKho6EtUH jQ0BGK1bbxSLHcinRsb9eNmr/iEXnf5uJ6UtxHWhvB677LT+/wAX5/zcHSluDrQ3g9d9l/8Afi87 /NwdKW4OtDeJXrTZVkdxuMdX01BL0GkUy7uDpS3B1obxXrxs3+YRfn/NwdJ7g60N4mTrXZnjeM7h EA6lSQWqKimVVwdKW4OtDeSYupbea25sE6zRGqcxTlqUCoOQzz8mFcKbSyMk1VGp536r+d8XV+Su OD/ZzOx9ORkfac++G3/Ux8tsdrSPg+Jw9X/RcDze13X0LZ7ZNWnXLO3m0DF3cypPkHZxrDmW217h CNw3VH2Z92aeRJAY41cxhVZmqSrUBXOnbTEzTcYtSy4FcWlOScc2JKmXbLpbC2h6Su4GMkZDqjMz wo+qbujSSWzGZwqc1VuaJag6JQZFvL3ZFkt1telriCWKR5LkvqcvEUcBVBBUULBq+5hoqbTrNCyc E1SDDjvumDFeKnTVxLcm5Ohe+fRoqJ3Goc2OmTIjL8WBq5h71s9Qrbx9r2k57zp26j1N0fdRRwKI 10BhU6+DHu56VPeNanLFdLi+6LK239GZ/eY47m6f7N2O529KRhbcpI7LQMGqTn3iV7OzGm1Jpe6S ZmuxTfti0VLWt4tsLt4JFtWYIs5RghY1oAxFCe6fNi7Oq0riUZXStMCwtN1FntKDVTXNKfMiYx9z KkuR0OzVYvie0879SnP/APk11/3K489X+Tmd76cjP+0598Nv+pj5bY7ekfB8Thav+i4Hg3V+4va2 m1ovB/ST+PUgxXqE6XFwNGmwrbfE9N6Dtup719wGz39tZI3I5wuQjh3aFioAcGgpqBPu4suSgoQz JvAqhGbnPK0sS8lg8SYp3dd6252VucxIQUq5KkqY+3yDhwxUnZ3SHavb0OX36R/StvjO82Ja/rBJ cRIKhgjSMzdzPuj4HEjgMEejR4PAl9aqxWJWQW/W2veYo7rbo5Hkha/u/eqWkjLnSQvLq60D5Z9n lw7dr2/ISl3HYWUsHijPdRD7UsIp21qCoCh3RSzM1Y9NQoA8i8MjhE7CWxjtXm9qHJ5vE+4uDyt4 21StESGIDSTGApA1RkmmrtNM8sQuglskDd5vbEp946O6wnsbTbLndbKa1tgula6FiZTIAoYLVqam Fe38WV1vuLak5JOpVcsXJRUW1Q8p6puZLfbbchwpSSbUvEmugccJ30sa+Rbp8cKeZ0Tzf3eOb/pm qv8AU1xxa+47FMCt9pz74bf9THy2x3dI+D4nA1f9FwOeevVLQ7OBxPpHu/CTGTVf1XA2aT+T4jcf WVyNpSaW3tZ500xtqDhmCjSCwWRc6AZ0wkO7uRtKjWGFKItl2luVx1TxG167uOaobbrQhgC1OfWh FT/OYVahdrtXoS9PtefqJPXlxpqNutDmQP5b/uYh6hd3r0J/59rz9R9+tb9IolawtTFKWYRgz5Mg pmpkArT8mGfeXVR1WPkKuytOqxw8xp/EG+aKJWsLZ0TUI0LTnTU1NBzMq4T/AH3Vjh6DvsLfn6hN 11dAtq2u1QgVAPOHb/xMM9Qu+XoKtPtefqKfridYon+z7UmStV+myoaD+cwPv7tFivQP8FqrwfqQ t431d0tJVuYkhktm020UIYAaiNROotUnT2n8GEuXupFufy8Cy3Y6bSh8fE6nqf2ba9v2Rx/qMYjW Qvac++G3/Ux8tseg0j4Piee1f9FwOeuvK8rZqCv944/xk8uMuq/quBs0r8nxMov+HEcrvGQ6ZtIz yFVrx93HP+mzx2nR++3wFW1jK06a1OkgVDGhzXBC264hKWAzMAsVAQe+2WWWXnwj2DIlXIswluUY iejc5OIBp5DkM/dxbNRoqbfErg5ZnXYQoxRoycxq4ebFMcGi1kq4dTCyawzJq4fGatM8WTeAkUIn KeiW30JRs6y0ybPy9uIlTKsCVteI8k8g2+9RApiZ11EmjZnsUZdn4sWRm8kkthXKKzxb2nW9P3ba f6R/YYzF5C9pz74bf9THy2x6DSPg+J57V/0XA556+NINnPGnpGQ/jpjJqv6rgbNJ/N8TNRwzptfp KuNAlqFoahuFa/i4YxqLVvNXCpuzJzp5Dtq0zqWrXStdRpQd3DQbaJlRFfKkka8uTiGqe3MgYzST WDLU6jjpbssJQnmMG5qnsI4cBllhmo0VNoqbq9w0B3Y2OdWNeHZTCjD1wiq0jIKIwBpXPM4aapsF iwTPI1pbqWBSrAClCCD2nt44JN5UCWLEaJFhn06uWrKGIHd4mla4ijo9xOFUdg1/dur/AKT/AGGF GMx7XHUt/tXXG2Q20Vu6vYBiZohI1eYwyJONNjuLkFSLoZr3b25uslU8Bv8Aq2+v1iW7tLOQQauU ORppqIJ4MPJiLtyVx1k6sm1bjbVIqhE+1l/y+z/5R+dirKW5hyLeyG0LY2KiSiMTCaUJpU97DJtE UqandumEsUVvTdiuQWRSIatpLy8qrd7IAd4nyUwu0ZEh+kpuTBrvuntJICRAjmqrHMuobs7QTiXJ tJbhUknXeUvUW3fYkNtI/wBj3vpBNFtF5hSiq3fo2RzpTC0GqUh3gHP0Czzy/kj87DNVIqbLpLo7 1i2o7gJ9qsgsjxtBLBKzgIK6+6TlhWiamTud1a3lubUWdk6rIyOyxGjGNiAwq1cOng1vFe06t5n7 unM0L/herRTufyHCnkxV4jifaR8GOp+tepLLd9ruLaO3t7UW7RS8zmFw7MT3VZaUby4lOgNHjn7M nXv9Pa+ab5mJzkZQv2ZOvf6e2803zMGcMpqLHwo8WrK2S3hO1GNI0j79vI7EIoUVZlJ+COGWIzE0 Ft4XeL5uHnV9rR2cPQW8hUDUWKhWVhpYsa4MwUFHww8WzfXF7p2fnXPJ5gFtIEHo6FE0rpotQ3ep xwVCg/B4e+McF5JdKNnZ5GZ9DWshjXXSoRNNFFVqKcDgqFDO774B+Jm93MdzfTWPPjTl64o5U1DW z1YBKVq/mxOYihBT2bPESMUjvIUHkUzqPyLgzBQQfZk6+Nfp7ap7aTfMwZwynQnqPvP6GvVmi+ne heia6No1cvRq4aqV9zCjHr97wxAEPAAMAAwADAAMAAwADAAMAAwASf8Ax8AH/9k= ------=_NextPart_000_0000_01CB3ACA.1C64E450 Content-Type: image/jpeg Content-Transfer-Encoding: base64 Content-Location: http://www.stata.com/distrib/policies/webforms/vgsg2_thumb.jpg /9j/4AAQSkZJRgABAgAAZABkAAD/7AARRHVja3kAAQAEAAAAPAAA/+4ADkFkb2JlAGTAAAAAAf/b AIQABgQEBAUEBgUFBgkGBQYJCwgGBggLDAoKCwoKDBAMDAwMDAwQDA4PEA8ODBMTFBQTExwbGxsc Hx8fHx8fHx8fHwEHBwcNDA0YEBAYGhURFRofHx8fHx8fHx8fHx8fHx8fHx8fHx8fHx8fHx8fHx8f Hx8fHx8fHx8fHx8fHx8fHx8f/8AAEQgAaQBUAwERAAIRAQMRAf/EAJoAAAICAwEAAAAAAAAAAAAA AAQFAwYAAQcCAQADAQEBAAAAAAAAAAAAAAAAAQIDBAUQAAIBAgMFAwUMBwUJAAAAAAECAxEEABIF ITETBgdBURRhsSIyQnGBkdHSI3MVNRYIGKFScsJDUxfBorIzJuFigmODRGR0RhEBAAEEAQQDAQEA AAAAAAAAAAERMQISAyFBEwRRgUIiMv/aAAwDAQACEQMRAD8Ac/iU6mc88q826fZ8v6lLZ201mJJY 41DAvxGFdoPYMVEJylZvw8c7818z8m319rt7Jd3cWoPDHI4ykRiKNgNgHaxxcYwxzzmHT5Lq9EZI kO40w9YRPJkCbUtRX+KcPSGfly+WhquoZAeKa7cKcIPy5F+oa3rEYXh3DLU0PwYcYwU82XyXHmrW 4zRrljt8mHpBRzZfKSfmfWZhktrllcb92J1hflyLX13nQXDEXp4SkEio3UqezFa4lPJl8jYee53L DxUhopPq4fjgeafkw++1190frKvz1Muenb30xhTq66/yV9TnYa1DQ0+ZHnONOOzm9ieo3kFmOjTG tT4hht/ZXFSjGZotAV2jFaUO/CX2LZrnS8l0/jIeHYkrevxEpAyrmIlNfQIXac3ZgiUZYA1v7Oa7 ks7S6hmuIhmeBHV5FUgGrKpqB6Q+HDZz0eJbaSU0cVA2jKKbcCbkeoPpsUNzM9zGkVoSLqRnULEy gEiQn1TtG/DqKArK6j4XjvEQ+CKcXxQIMWTfmz1y08uCYXjJkXgdni4qtcNHxDEpGbhn0VfLvynv wlyrEsLQKTH6zVBzbRSlcaMzOrf02r7X6K45f07/AM/Rt1Q+2oPoR5zjTjs5/ZvBr02FdEuP/Zb/ AALgyuOGyxyoRLmI2Ag09zDg8o6uZ8jdP+YbC55lXXpbae05qt5ZL9IAax3c0kwYHN/mDgzKM+z1 QKYVD2hU16P9QItG9KeF9TnibxvAuWhR5UnThez6S8CFcwNO6u7AU0Q6lyXz+1zdaZNe2zTzwyzL B4+YO8Mk8TNIiH0lWJlKguThorEPWo6Dd6Ryldabrl/axavqepWV7arPMQt01qlqsiNIVNC7xH0m FNorvwJrWehLLyfzjNpTypfWgseDNxp4bsJCjQx3UUiFI0RPRaVS7AAVU7K4dFbQFn5B5uuZbi5t 7hZPFZW8RFdvmuYWklkiIah4SqsiUA2HLgocZw7Xa2rLY28ctXmiiUSyMPWZUozVI21OHCZAU/0T T/nU/vYw7uz8/Qzqh9tQfQjznGnHZzezeDbprDLJolwUYKBctWv7C4MppI4cZmFsawmbe42+7id4 a+OXkabKu0MK+6cG8DxSjmtJI0zMwIrSgr24cZRKcsKQ4x1Q0ex1Hm3TB4u0iv7i0FqLO5SciRVv kljWeWJHWOGfhtFVyNvfhoiKQ8avyXfzaNyzY2Os6e2o6LbtHNfvORKq5kzvHQsJYgEKPHKMp2bQ RimdesqpJyAv1v4IarpiQWdveXXDgLmSS2uzd0YAKRL6+1EbZk3HZRUXssHI/JFxo2ttrV9qOnJZ rp8Vu0EEpQrnjh4JkWQIV/y2pm27dgGHAy6wvU2o3CqyqFIAO2mLoyqX52+4mf2uJXyVrjm7u/8A P0P6ofbUH0I85xpx2c3s3g16bXLQaLP6OdGuWAVQS+bIlNwpTv7sTyXX6/8AlYzc6hcMeGzBe2OF VoAdlDK5ysf2d3lxnV06vUd3eWrcOVS6+ysjKG/4ZCaP79Dh1EwKuJBLaJIAQGINDvB7RisbsuSz mXMvIGoa5rSapYasNOUpDBeR5JC7La3LzLkZHQUcSMjq4Ipi6MYyhUNf6GaiNKij0u7tnax0vwVv arCIONMYHhkZnBpSRpM5z1PZh0LdsdDZSrzTajBHPKS/zVuVSIu1wzpCM1VQ+K3V9ny4KJ8jdz0b uH0+4sk1GzJkdZYLl7VvEB6bc8oerKn8NfVpsIwan5IXj6kuo7KGAkSNFGsbOKJmKJSuUbBWm7F1 ZTBfwm+4vC9riZaeWtMc/d3/AJ+h3VD7ag+hHnONOOzm9m8G/TS1jm0S4LlqC5Pog0HqL3d+J5I6 tPWmmK7qqqAqgBRsAGwDENkN6iNbtnUN3VAO/BEFMzEBIrDiIJA9N4AJPf7uLrRGszFatppCoCFY AEkmgO8+/g3LxNnSgRQsPgPx4Ny8SC502KJAWAYE0pt+PFY5VTlxxCvW4kkjV9m/+3F1c0wPkibK c1NgNKe5hNVRp/pGn/kfvYw7uv8AP0I6ofbUH0I85xrx2c3s3hLyRrg07SJIykjcSdmqihhsVRtq Rh5RWU8edIWVOai26Gfb/uL8eJ1hp5ZSnW5pssZhlUOwBLKoA29tDgiIOc5keuqWVrbqJ5VRtpK1 GYDad2/E5R1a4WQS8z6YrZYpRI2zKqq7MxrtAAHdhareH1zUZYgbTTJi7GgM2WEDtG8tgoTUb665 kbUY4ooSQYI4znYGm0M3+zFY0Z8lg9hEDZoabwduKhjMA51KoxYmlDTF1c6rf/Hf9f8Aexz93ofj 6FdUPtqD6Eec4147Of2bwF5Zn4en075XP6FxbGLLJbXINMANI5M4A7iPPiZaYio9C0+5fxU4aR2o MpYhRl2bhjPKerqwsYW9nbW5+YjEY/VUAD9GJXVq7FICx2tWmam2hNaYeN052CIIZyI2lbNt2EHL mG8AnYTh7UTHHMxV6SwtlbgrKoYCvDFK09yuDcvEyTSISjVI3H2fJh7l4Yc94Y+7XD7PF5f79MRV tTpRJ1Q+2oPoR5zjXjs5PZvBNpU4jtEBNKu5/wAOLljFlis3kdaptFMKp0NrdjCodwTShpXfgOOg uHXbkKFjt5AvYKriJhrHIlGq6goAFpIASTQFN52nBSFby39Z3krpFJC6Kx2klaDZXswREFOc2NUh SeyRHFNlVYesrDcy+UYzyu3wmkPQsLXhCN0D7aszesW7WJHbgVtKKS3niBS1cJGymuer5CP1QT24 R1ju51lP3cy5jm8XTPsrXPvwB76ofbUH0I85xtx2cfs3hUzccKCAA0q0nnXFTdljZb9AmZ7CJ9lG rt9+mJlcWNrmYLHHtpVqfow4Tmltrk5gooQCN+GiJO0ZmO3ZTuxLoRyj5wbe0ebDKbg7nmG+tBJB Fa8TIaJJRqU94YzmOrfGejSc1XxGVrM5gd4DUI392FRVWXPMl8YFmjtSpXMjoQxBqta7vJgoVVO4 8n3P8Rl+c8RxMtO3NWlMSvsJ6ofbUH0I85xtx2cns3go0lImsVzorESPQsAexe/DyujCyvcxy8wj WKaPeR2sUdRLFICUJIUggL3bcAmYHcvrz7czqr6raiMAmUCEliKEDKWB9ojswpEUPIdJ6nqVki1u xooAZJIMytRQCxoqmtduBXRatKlvo7KFdQmWW7CATyxgqjP2lV7MUxiTI3A4QZRmYGoG6vkrhNYy eU1PVQAq2xAG4cQfFhTELjOXsapqlaGE17uIPiwtT3lt9Q1VkbLAdx/ieT3MFIG8qTmk+6uenzni a0r7Wfvxn3dHZz38TnPvMfLvOGn2ulvAsMtmJHEsKynNnYbC3uYrGUcmETPVyOLrl1EiQJHc2qqC SB4WLed/mxVUxjEIJesvPUsrTSTWrSP6zeFj202YCnCD/k/qr1H1K/mt7bWtL0to4TIZryCJEYBg Mikg+ltrTBURxwfwdTurTyNGObNEQLUEtDACfcFNtRtHeMKp+OA1t1V6tyT2sTcyaNAJ7q4tGeSG CkJt1JEstAaRy+ww34dS8WIpOq3Vg1ROb9F9ByjFraJVB8jEba9mFU/HiWa3126yaQ8QfW9PuUmL BJILW3YehlO2gNK5sOo1hNyp1w6wcw6jJZrr9hYlIjLxZ7KHK1CBlFBv24KnGMI+Z+vPWDQdUk07 6/sb0pGrNNDZQhatWq0Za1FMIUh1X671H+if15nX6y8J4ziZRl43Dz5sm6mbsxDXsF/EV0h5m5w5 is9W0ue3S3t7UQNFLxOIXDs1RlVlpRsETQpirkH5dOfP5lt8EvyMPYtWfl058/mW3wS/IwbDVh/D nz2d8lsfel+Rg2GrX5cuev17X4JfkYew0Z+XPnr+ZbfBL8jC2GrPy589fzLb4JfkYew1bH4c+exu kth70vyMGw1bH4dufV2rNbg+TjDzLhbDVjfh15+bfNbknZU8Y/u4Nhq7z9ytZ/o993KL47wnhc9G yZuHkzbq0riVOu3u7AYPAGYAzAGYAzAGYAzAGYAzABP/AG+AP//Z ------=_NextPart_000_0000_01CB3ACA.1C64E450 Content-Type: image/jpeg Content-Transfer-Encoding: base64 Content-Location: http://www.stata.com/distrib/policies/webforms/wdaus_thumb.jpg /9j/4AAQSkZJRgABAgAAZABkAAD/7AARRHVja3kAAQAEAAAAPAAA/+4ADkFkb2JlAGTAAAAAAf/b AIQABgQEBAUEBgUFBgkGBQYJCwgGBggLDAoKCwoKDBAMDAwMDAwQDA4PEA8ODBMTFBQTExwbGxsc Hx8fHx8fHx8fHwEHBwcNDA0YEBAYGhURFRofHx8fHx8fHx8fHx8fHx8fHx8fHx8fHx8fHx8fHx8f Hx8fHx8fHx8fHx8fHx8fHx8f/8AAEQgAaQBUAwERAAIRAQMRAf/EAKkAAAEFAQEAAAAAAAAAAAAA AAUAAgMEBgEHAQACAwEBAAAAAAAAAAAAAAAABAECAwUGEAACAQMCAwQDDgMGBwAAAAABAgMRBAUA EiETBjFBIgdRYRRxkTKy0iNTcxU1FhcIGKGxQoHBUsIzJPDR4WJygkQRAAIBAgQCBwcCBgMAAAAA AAABAhEDITESBEEFUXGBwSIyE/BhsUKCQwaRFKHhUmIjJHJjRP/aAAwDAQACEQMRAD8A0P6pPMTr npPLYKHprITWcV1byvcLEu4MyuApPA6lEMm/T15h9a9R9NZm66gv5bq4t7xY7d5V2kIYQxA4D+rW kYpi96bieqxZm/aJGMxqQCdWcEYq9KmZHNnMgsqgTGhWv8dCgiHelXMac7kaf6x1OhEevLpOLnsi VB5zaNCBX5dI6POZEzBTMaUJpocESr8q5hO0yN1sMtyz+zkgc0Gm0n+7VHFcDaNx5vIvTJdPb7ra Zt4FRU1DD3dUVOJo60wAz3+aZm5bPSJWMvdTbrTShd3Jg38aX34Z9rr/ALndyuZ/3VpXWdMRqvgr 7jL+e80seSxex2WsL1oSP6tdLlyTTPJ/k85KUKNrBlDy5mumwt6d+75/jvJJpyx2atu0tSK8knJ2 pVdfF3GshkuuUm0ptp4a1rTSjodpN0Ip5ZxKu8rWnCnZTUpFZN1GmWWh4jU0IqziyybBQilOFdFA TZr7K0xZt4HZITKY1LEkVqQK9+l22PwjGiyLJtsbtK7Ydp4kVFP56irL6Y+4ntliSBUg2iECiBTU U9WoZaNEsAH1T7MixuH2XjAqFXsdO/frS3UX3NF1mIqfwZXv9o/zaz+YZ+32Efn195Yr6mT42uly 7Jnk/wAp88Opgzy7mSPCXoZgKzmgP1Y1bdrxIpyN0tS/5dxq7e6iEEYLqCFAIJGlWjtRkqENzcxm daMDRTUg+vUpFJSVSM3CUPiHv6mhGo5HOmweIdmihCkPhuIxcKSwA2kV1DRaMlUK2H2fNKfarmOG BfhVYBm9Q1SVVkbw0t4uiDs2fxFnZKtpLFIVG2KJCKCnp9WslBt4jLvwisGZLIX5nuVleQO7BizV 9PZreMaIQuXKuoM3D8Ebu7n1r/7aV+btOt9r6Rvn195Yr6mT42uly7JnlPynzw6mDvLqOF8Ffl0D ETHaSOI+aHZq278yKckS9GXX3BbF9Q9J3EyY5Lu3bIr83JA4KvzFALL4wAWoQaA6UcsTuQt+FOhz LZfp+yvLYz3NvDFcRsYm3DY5DqnAiqnxOBqVLApK23LBDZcx0+tgb03VsLUqCJd608YG3h21O4am pGh5UIrHPdOTTNZJeWz3sCg3Ee5arup2sfD/AFjv79DliQrb0p0L8EuOa+WrwmNYnd/Eu0BaeJuN KAHQ3gRGPiyGXvUXR9tBcvPeWqJbSLbzuPEEllj5iISoPEp4uHdqmr3jHpV4D7LJdPC2tVlubTnT ozLukjq2wBn4k9ysDq0pYmduHhyJb2O2Fzb7EUK6uTtAoaDgeGpjkUmlVFCi/gmlPDz+zupu0p83 adf7X0jfPr7yxX1MnxtdLl2TPKflPnh1MC9Cvdrgr7kBCpmO/cTX/THZ3dmtN1TUjDkzfoyp09wL vemsNnbi7SXKRRzxXKPPEXVHillghWNAWHwjHACKceJ0lKMW2d+1duRiqJUoUr3orBT2Fh02+Wj3 WKz3MU0UiCqzFo2JPFao/GnpGjRFqlQd+6p6qZkI8vOnZ5bm3XJRytNGY44laOoE1SOW9PG/hahB qBw7tHpxB7m6lWhVj6N6XvIWdM0m6JKuCqKdiryvgbRubbA3Zxrx0OEeklX7q+UJ4noTCW8t9bWe SRzmITbgNLGwQFxLtUHiW4Iu0jiBqVbjR4lXuLjksMjtrg+n0xU+Ds87BFOt4bh5TtV+bbwGCVF5 irGyhd3Edh4aqoxyqayuXa10rIrY/wAuMNfxzRWOaiZ3do2jrE8ihUAaPbSvBSKkaHbjXMrHcXKJ 6V/E3cNte4+3xuP3I62duIIXqfEsahSW9dBrWKVBS7OTlV0JKv8AgOvDfzf7K10n8/adv7P0nfPr 7yxX1MnxtdHl2TPK/lPnh1MC9DTxRYK+3uq1lagYgV+aGtNyvEhflEkrMuvuBv5bdO3+69nyM/Ou yZrhC8ZXmNu2lVYeExhyBTScrWLO9Z3lIJYYIG3flhhYr61SHINMvM5100uw1UMSUXb4av314aFZ wInv8ae4s2Pl5hLC5tbuPISNLZOssdRFQ7CxCigqB4u7VlZoyk962msMSJfLnCXG2d8hKrEsyr81 VFcsTGCRX4Uh4nRKziEN89KyOW/l3jbbqDGXNlfKot5VnuDMFO7lSCQBSoFGYjt1Ds8S0d9V0fFB bJeXeEu5Ly4ky0qtcySTlRydq8wHcAtKVNeLdpoNV9I0W8ouGBH0x0Bg8fd43Kx5CQy2dXS3flBd 7R8pqkANxGpdqjKLeaoY0xNRkbyA3tsBKposlaEGlaU1rGLoKXLi1LE5uH4A3VFObWvd26R+ftO/ X/D9I7z6+8sV9TJ8bXR5dkzy35T54dTMr0wkTYe4LqGIlahI4j5tezW9/wAyOfy2npPr7jZ9JxdO TQQQZSxjJnA5F424Bm71Y1p/bpS7qWTO3s1aaSnHPiauXovpYv8Ad6Dh3M4/v0ur0+k6T2Vn+kwX U/TU2FnJ28yykJ5E9OIPbsf0N/PTdq5q6zjbvbO0/wC0C268xAEj3sF3MANxoO0mndrVikXUktBC 92oZQV2Mad1RTUPItBpyCDw2uxvmk7D3DVMTdpdBDZRWzWkTNGrMy1JIBJOplWpW3TSiDILAksWx FXcGqAKA01MSlyiaCfD8svVXs1zvudp6j/zfR3Fnz6+8sV9TJ8bT/Lsmea/KfPDqYM8u8BdZTDXb W/L3RXG0mVyAN0YPwQrbuzv1purijJV6Bfk+2lctOlPN3Gxu+n+qp8THj1XHCMPzHkAaM17AFVIw F9Z79KK5BOuJ2Z7a+4KPg9uwLdPR9Q2cXs+Va3mWMUt5Y2Zm2/4X3qK07jrO5peMRraq7BUuUfQE sikGQgkt7qJZIJF2PGe8e76fRqkcHVDF1KapJYHnF2Mv0ZeJybaK9x8hb2Od0HM3uKBJHCluHoHw tOx03V0M89cdzZywSlDg+PU/bEA2SzPfc6ZVMrhmMUZ2oNx48QD38BQeoV7tpZCVuTcqsKNJZsrR pETMAdw51RwHdROPud3r1nRjLuRySx6/5FCNLxbSM2qo8e3wq7eKnaKbQQeH/SurulcTBOWlUBeQ u7vnRcxUHhbbtJIPpqdaRiqC129KqqaDcfyr3d9K+rXJ+72ntK/6lf8Ar7i959feWK+pk+Np7l2T PO/lPnh1Mi8qc5iMZhL/ANvvI7UvcBkEhpUCMA01O8tylJUXAryLc27dqWuSj4u43lt1f0xNGDHk 7duFTRu7s7xpN2Zrgd2G/sNYTRPDk8deyMtlcJOYxWQJU7amnHhqri1maRvQm/C60LAjlYVVCQe8 AnUVNKMynVucRkOCsYxd31z4JVA3bKiu1R3v7nH+emLNv5ngjk8w3OHpQWqT9v1AGC6WvJsg1vkb W8sw0Z5d4ioYge9W9ZH9Q4f0jhre5eSVU0xDa7CUpOM1KPv4e3v7AVm8ZdYa6e3uyFIBaKYGiuo7 GU/8U1pbmpKqE9zYlZlpkQYjJQy2UKzSqkm0AsSACa9/ZQ1/jxFD2zOOJSxdTiqvEWXw11ePFNZo ZWUMXCeIGvfVeAP8D6tELiWYX9rKdHFVCfIn/KvkctufTZy6eLd2Up6dcuv+Wv8Acey0v9nSmPp9 xd8+vvLFfUyfG09y7Jnnvynzw6mY7BWdpcYS6M8SyMjybGI4j5tTwOmbsmpI5OztxladVxfwND0h 0rDlIbTk26KRGjz3RWu3cKk17ye4awvXtNTo7HZK4lRcMWeuYrB2NlaJBaRrFCtaqBQs3Zuanfrm zuNvE9XZ28YRpFURYldrbwQ0oT8528BTu9eoWJdvTgjAdaYNLZPt7HulncIwacAhWd2/qRv8XpHf pyxcr4XicLmG20r1YPS+Pt0gLp3qqaLMq2SuLu7pE7Q24kAhJBA3SA8T28Na3LVY4UQltN61c8bl LDKuHaMzrx5ieS5vokkfaVjWnCNe0Kvo1Nvw4Ijcv1nWSA+HxmNfF20kltG7ugZmYVJJ1pObq8RS xYhoWCK+UsrOC+tRBEIlkSTeqVUHbSlQPd1aMm0zO9aipKioH6D8rqd1Ncn7naezp/qfR3Fzz6+8 sV9TJ8bT/Lsmed/KfPDqZisRJfpg7n2WFJAXk3l22kfNr2CnHh69M3aalU5OzcvSdFxfwDfT+f6w tcPbQ2SqluEBQLJtJFOBYBG4kevWFyEHLE6O23F+NtKOXt7g3ezeYkuPXKWOTE8UaEXNjGWWRO9m JK8dvq7tZRVutGhu4904a4yr0riVuk+oMTjI2vcq9xPkKNsjTdJHGvpBYjcx9OrXbblgsjLZbi3b 8U23L4AXM9U5zqG4jlFuq2taWVtzDUBjQEgDi7a1hajBCm43dy/JOmHBe3E5JgOrrCW2yUtiixMH hdGYgoTQ+Oo4Vp4aV0epB4VB7W/BqbjhkPmuM7yZNtpBXaafOk93o2jQlHpLSdymS/Uq4afM/Zds EtoWjCUjZpCpK14EgA6mdKmdh3NCwX6lXLT5Y5C15lvEpCSbArlga03VagpTh3atClGZXnPUqpGm rJ+VVaDmba7a8K+iuuV9ztPaY/tPo7jO/q16szGCzOATHmILPbzNJzYlkNQ4ApXs1Fq9KHlZbdbG zfadyOqh4VB5x9dQQNBDPbJE5JdRbR8SRQ9vq1o9xN4tmMOWbeKoo4dbCOJ/UJ5pYmAwY+/toIiF G32SFvgig+EDrOc3LMZs7a3bVIqiLp/U55xlg32rbVHAf7K3+TqprpRNB191HdvePe9UYy3lWRBw tYisvNQM7IQyrRGYqfc4ejWivz6RN8tsN1cfiRt15l4riK0teprF41qHmezjRFVTH8E7zXg7U/8A E6n9xPpIXK9uvl/iyTqDzQ6uxtjatY9VWGWVHZVtobVF5YkFWdlJ4mq0r/HUetI0/YWaJUwXvYBP nf5hkEG6tqHgf9rF/wAtHrS6SP2Fn+n4hHpfzP6/yVyMdFk7GxihiLo81tGVopACj1muh3p9JEeX WEqKPxI+p/MvrrHZKOCTJWV7JHHuEsVqihd/apB76AHUq/PpIly2w3jH4ntf2te/t6+1ty+2+we0 7to28zl7q7eyle7WGp1qO+nHRp+WlAp+pXyk6k64vsRfYme3jisIZIpY5uZvZncMCuxWFKDv1CZd o8P/AG19ffTWvvTfI1OojSL9tfX301r703yNGoNIv219ffTWvvTfI0ag0i/bX179La+9N8jRqDSL 9tfXv01r703yNGoNIv219ffTWvvTfI0ag0i/bX199Na+9N8jRqDSdH6bOvx2T2w9znD/ACaNQaRH 9NvX7cWntifSecf8mjUGk90/AGd/JP8ACvg+0PY/ZOZ4uXv5e3d2bttfVqpY9svfgaAKGgBaAFoA WgBaAFoAWgBaAFoAtf8Az6AP/9k= ------=_NextPart_000_0000_01CB3ACA.1C64E450--