Università degli Studi di Roma "Tor Vergata"

The first part of the course will familiarize students with the workhorses of empirical work in economics, namely the linear regression model and the ordinary least squares (OLS) estimator, and with the problems that arise when the model assumptions are violated, in particular when the regressors cannot be regarded as exogenous, that is, uncorrelated with the regression errors. The second part of the course will pay special attention to the method of instrumental variables as a way of solving the endogeneity problem and the specific issues that arise with the use of this method.
Programme
i) Static Regression. Introduction and review. The classical linear model and the OLS estimator. Sampling properties of OLS. GLS and feasible GLS. Diagnostic procedures. Hypothesis testing and model selection. ii) Instrumental Variables. The instrumental variables (IV) method. Estimation of causal effects. Properties of conventional IV estimators under weak instruments. Robust inference under weak instruments. The generalized method of moments (GMM). Weak identification and robust inference in GMM.