Module Title:
Module Code:

Module Content
  • The Single Equation Linear Model.
  • Structural Models. Sources of endogeneity: omitted variables, measurement error, simultaneity. Asymptotic properties of OLS: consistency and asymptotic efficiency. Heteroscedasticity-robust inference.
  • Instrumental Variables Endogeneity.
  • Reduced form equations. Exclusion restrictions. Rank condition. Two-stage least squares. Consistency and other asymptotic properties. Potential pitfalls. Local Average Treatment Effects.
  • System Estimation.
  • Seemingly unrelated regressions. Pooled panels. System OLS estimation. Asymptotic properties. Generalized Least Squares.
  • Linear Unobserved Effects Panel Data Models. The omitted variable problem.
  • Assumptions about the unobserved effects. Between and within variation. Random and Fixed effects. Consistency. The Hausman Test.
  • Maximum Likelihood.
  • Modelling the conditional density function. Likelihood function. Consistency and other properties. Hypothesis and specification tests.
  • Discrete Response Models. Linear probability model. Latent variable models: the Probit and the Logit. Interpretation. Marginal Effects. Tests of hypothesis. Endogeneity problems. Unobserved effects probit. Multinomial response models. Ordered response models.
  • Generalized Method of Moments.
  • General and optimal weighting matrices. Esti-mation under orthogonally conditions. Minimum distance estimation. Tests.
  • Corner Solutions, Censored Regression and Sample Selection Models. Top coding and data censoring. The Tobit. Expected values. The Inverse Mills ratio. Reporting results. Specification problems. Unobserved effects Tobit. Selected samples. Truncated regression. Heckman's model of selection.
  • Estimating Average Treatment Effects Counterfactuals and self-selection. Methods to control. Control for selection: regression versus matching. Differences in differences and regression discontinuity.
  • Extensions to Panel Data Models. Unobserved effects without exogeneity. Models with individual specific slopes. Hausman-Taylor models. Dynamic Models