KECD 2235 - ECONOMETRICS 2 : Estimation and inference

Course description:

The course follows up on the course taught in the first semester on probability and statistics.

It introduces advanced econometric techniques used in empirical economics with cross-sectional data.

The course discusses identification and inference in linear, nonlinear, and semiparametric settings, under standard and

nonstandard conditions.

Prerequisites

Students should have a robust knowledge of probability and statistical theory, covered in Econometrics 1. Some

knowledge of R might be ideal.

Organization

This course is organized through lectures, practical classes, problem sets, office hours, and individual and

collaborative student work.

Assessment

Final Exam (60%) + Midterm Exam and Problem Sets 40%

Additional information: (outline, a few bibliographical references)

Outline

• Inference in linear models, identification and causal effects, OLS estimates, measurement error, control

variables, testing.

• Inference with structural linear equations, Two-stage least square estimates, specification tests (endogeneity,

functional form, heteroskedasticity).

• Inference with a multivariate system of linear equations based on OLS, GLS, and FLS; Seemingly unrelated

systems of equations; the generalized method of moments (2SLS and 3SLS), optimal instruments.

• Inference with nonlinear models: Nonlinear regression, maximum likelihood, quantile regression, minimum

distance, M and Z estimators; Wald, LM, and LR tests for general models; Numerical optimization methods; GMM

under general settings, asymptotic theory.

• Causal Inference with Machine Learning.