Statistics, Quantitative Methods I

Models with discrete dependent variables and applications of panel data methods in all fields of economics have become increasingly important. This course starts with a brief review of concepts previously carried out in other courses of the program and focuses afterwards mainly on the methodological and empirical issues concerning the analysis of cross section and panel data in the specific context of economic models. Selected topics in time series analysis, especially topics of importance for the panel data analysis of dynamic models, will also be discussed. On satisfactory completion of this course, students will be provided with a number of sophisticated econometric tools which are of use in advanced empirical research or professional work.

Chapter1. Endogeneity of Regressors 1.1 Stochastic regressors and the properties of OLS estimators 1.2 Measurement errors in the variables 1.3 Simultaneous equation bias 1.4 Instrumental variables 1.5 Testing for endogeneity Chapter 2. Models with Discrete Dependent Variables 2.1 Models for binary choice 2.2 Estimation and inference in binary choice models 2.3 Multinomial models 2.4 A Poisson model for count data Chapter 3. Limited Dependent Variable Models 3.1 The truncated regression model 3.2 The censored regression model Chapter 4. Panel Data 4.1 Basic panel data models 4.2 Estimation and testing methods for random and fixed effect models 4.3 Estimating Standard Errors in Finance Panel Data Sets

The course is organized in theoretical classes whose materials are slides and problem sets for consolidation of concepts. In addition there are several laboratory sessions where students will use STATA.