Regression Analysis III: Advanced Methods - icpsr.umich.edu

1 1 Regression Analysis III: Advanced Methods Syllabus - Summer 2015 Instructor:Dave ArmstrongDepartment of Political ScienceUniversity of Wisconsin - Hours: 1:30PM-2:30PM M-F (or by appointment)Course website: Overview and Course ObjectivesThe Regression III course takes a considerably different form than the first two re-gression courses at the Summer Program. This course will hopefully prepare you for thethings you will encounter when you (attempt to) publish quantitative work with linearmodels.

2 Initial linear model classes focus on the assumptions and theoretical consider-ations of linear models and generally walk you through estimation and courses also deal with diagnostics, though these often get less time than theyshould. Further, it is not always obvious what violations of these assumptions will leadto in practical terms. This course will provide you with a systematic approach to assess-ing, fixing and presenting your linear model results. Though we focus almost exclusivelyon the linear model (we will allude to nonlinear models occasionally), the logic we followwill be helpful in dealing with nonlinear models as is a class that deals exclusively with observational data - those not collected inexperimentally controlled environments.

3 As such, we will spend little time on ANOVAand no time at all talking about concerns that are specific to the Analysis of RequirementsThis course is a practical, data-analytic extension of what you learned in your depart-ment s linear models class or the Regression II class at the ICPSR Summer Program. Assuch, I assume you are familiar with the types of things taught in these courses - Gauss-Markov assumptions, properties of OLS estimators, and statistical inference for linearmodel coefficients.

4 While I assume this knowledge exists, I will spend time reviewingthese ideas briefly in class. If you are not sure where you belong in the series of linearmodels courses at the Summer Program, please see me or the Summer Program directorand we will make sure you end up the most appropriate recently experimented with allowing participants to use either Stata or R and found,from student evaluations, and from both R and Stata users that the support for bothpieces of software was both distracting and unnecessary.

5 Stata users found that it wasreasonably easy to pick up R for the purposes of the course. Further, Some of the specificsoftware used in the course does not exist (or exist in the same useful way) in Stata. Ifound that the implementation in Stata often required some programming (loops, macros, ) and that was a threshold many participants did not want to cross. Thus, R willbe the only officially supported software. That said, I have Stata code for many of theexamples in class and would be happy to share with interested participants.

6 If you wantto use this course as an opportunity to strengthen your R skills, but have little familiaritywith that software, you should take the R workshop that I teach in the first three weeksof the first you re one of those glutton-for-punishment types, you may also find it usefulto learn LATEX. LATEXis a system for typesetting documents. People find it most use-ful for typesetting documents that are heavy on mathematical notation, but this is justthe tip of the iceberg. LATEXhas its own bibliographic software (BibTEX) and will au-tomatically build (and re-build) tables of contents, lists of figures and lists of tables.

7 Italso automatically numbers (and re-numbers when necessary) tables, figures and equa-tions, changing appropriately formed references to those objects when table, figure orequation numbers change. Best of all, common LATEX typesetting engines are free ( links to the software appropriate foryour OS). Everything I present in class is written in LATEX; specifically, the slides areall made with a package called Beamer . There are some nice literate programmingtools (Sweave and StatWeave) that integrate LATEX, R and Stata as well.

8 Further, thereare those who see LATEXas a sort of secret handshake for nerds. So, if you want to be oneof the cool kids, then you should definitely try it; everyone else is doing Course Text(s)No one text effectively presents all of the material that will be covered in this said, much of the material is covered (and covered well) in:Fox, John. (2015) Applied Regression Analysis and Generalized Linear Oaks, CA: Sage Publications, , John and Sanford Weisberg. (2011) An R Companion to Applied Oaks, CA: Sage Publications, R Companionis a great book for those currently learning R.

9 I would highly recom-mend getting the recently update and expanded second edition. This is widely recognizedas one of the best ways for Social Scientists to get into R. The Applied Regressionbook3is a great general purpose Regression book. Much of what we talk about will be coveredin other Regression books. If you ve got a particular favorite, then it might be worthsupplementing your reading from your chosen Regression book with pieces from the Foxbook that are not covered by your favorite.

10 Some books that I think are pretty good(depending on your orientation toward visualization, ) are:Gujarati, Damodar N. (2002) Basic York: McGraw , Jeffrey M. (2005) Introductory , OH: , R. Dennis and Sanford Weisberg. (1999) Applied Regression Including Computingand Graphics. New York: Wiley & Sons, will also use a number of other books and articles to deal with more specializedissues. These are listed below (along with the appropriate chapters/pages) for the classesin which we use SoftwareOne of R s main virtues from the grad-student point of view is that the base packageand all of the add-ons (called packages in R) are free.