The course provides an overview of some of the most widely used quantitative methods for the social sciences, with an emphasis on regression analysis. It builds on the course in Data Analysis, which is a prerequisite. Then the theory of estimation and inference is considered, followed by the analysis of several incarnations of the multiple regression model.
The course has an applied orientation. Examples draw heavily from political science, and are analyzed using the software R. At the end of the course, diligent students will be able to apply the methods considered, using R, and to correctly interpret the results of their analyses.
The main textbook is: Stock, James and Mark W. Watson. 2014. Introduction to Econometrics. 3rd updated edition. Pearson (indicated in what follows as [SW])
Also relevant is: Ray C. Fair. 2012. Predicting Presidential Elections and Other Things. 2nd. edition. Stanford University Press (The PDF file of the first edition is: avialable online ). A copy is available at the Central library.
We will use several "Shiny apps" - neat interactive tools to learn important concepts. They are listed in the following "course contents" section. Several of them are from the book: Agresti, Franklin and Klingenbergs. Statistics. The art & science of learning from data. Others are from: Metzger, Shawna K. (2020), Using Shiny to Teach Econometric Models.
The linear regression model with a single regressor
Esimation of the coefficients and test of hypothesis on the regression coefficient. [SW] Chapter 4 & 5, all sections - including 5.5. (the Gauss-Markov theorem) and 5.6 (Using the t-statistic in regression when the sample size n is small).
Shiny apps:
Linear regressions - Different datasets
R Lab 1: The linear regression model with a single regressor.
The linear multivariate regression model
The model and estimation of the regression coefficients. [SW] Chapter 6 and 9 (see also: 17, 18.1, 18.2, 18.4 and 18.5).
R Lab 2: The linear multivariate regression model, Part I.
Test of hypothesis in the linear multivariate regression model
T-test and F-test. [SW] Chapter 7 (see also: 18.3)
R Lab 3. The linear multivariate regression model, Part II.
Heteroskedasticity, testing, and robust standard error estimation. [SW] Chapter 7 all sections.
Analysis of the regression model. [SW] Chapter 9.
Instrumental Variable Estimation
Instrumental variable and 2SLS estimation. [SW] Chapter 12 (all sections) (also see: 18.7)
R Lab 5. IV estimation
Panel data models
Panel data structure.
"Pooled" and "Fixed Effects" models. [SW] Chapter 10 (all sections).
R Lab 6. Panel data models.
Discrete choice models
[SW] Chapter 11 (all sections).
R Lab 7. Estimation and analysis of Probit and Logit models.
Time series data and models
Logarithms. [SW] Chapter 8, Section 8.2, pp. 315-323. See also: Robert Nau, "Statistical forecasting: notes on regression and time series analysis: The logarithm transformation"
Time series models. [SW] Chapter 14 (Sections 1-6).
R Lab 8. Wrap-up of the course.
Students will have to complete a project using R, that will count as a midterm exam. The grade on the R project can't be refused. A "pass" grade on the final exam can be refused, and the final exam retaken, ONLY ONCE.
The deadline for the R Project (QM course) is Tuesday January 4, 2022, at 11:59 pm.
The final grade will be a weighted average of the result of the midterm exam (R Project) and of the final exam, with respective weights 0.4 and 0.6.
See at the bottom of this page. Virtually (Skype or Google Meet), by appointment.
Lucio Picci
15 July 2021