B3454 - BAYESIAN ECONOMETRICS

Learning outcomes

The goal of this course is to introduce students to the basic tools of Bayesian analysis, and to apply them to make inference in the linear regression model. By the end of the course students will: 1) be familiar with the main steps of Bayesian inference; 2) be able to elicit an appropriate prior distribution; 3) be able to build a posterior simulator; 4) be able to estimate classical and general linear regression models using Bayesian techniques.

Course contents

1. Introduction: Review of the classical linear regression model. Maximum likelihood estimation. The Bayesian approach to the classical linear regression model. Bayes formula. The likelihood principle.The James-Stein result.

2 Bayesian estimation of the CLRM: Theil mixed estimator. Prior selection via the marginal data density. The independent Normal-Inverse Gamma prior. Treatment of the error variance. Gibbs sampling. Convergence and mixing. The Natural .Conjugate prior. Marginal data density. The Normal-diffuse and Jeffreys priors.

3 The Generalized Linear Regression Model: Autocorrelation and Heteroskedasticity. Stochastic volatility models. Metropolis Hastings algorithms.

4 Linear and Gaussian state space models: Kalman Filter. Carter-Kohn algorithm. Models with time varying coefficients.

5 Non-linear non-Gaussian state space models. Sequential Monte Carlo methods.

Readings/Bibliography

John Geweke. Contemporary Bayesian Econometrics and Statistics. ISBN: 978-0-471-67932-5

Assessment methods

Written examination. The exam will be bases on closed-book questions which on the material covered in the lectures.

<18 fail

18-23 pass

24-27 merit

28-30 distinction

30 e lode:excellent

Teaching tools

Some Matlab code implementing the models will be made available and discussed.

Office hours

See the website of Andrea Carriero