Course description

Bayesian Econometrics

Introduction

Please note that this course will be revised before it is offered again.

Course content

1. Concepts for Bayesian Inference

  • Bayesian inference
  • Criteria for evaluating statistical procedures
  • Probability: objective or subjective

2. Numerical Methods for Bayesian Inference

  • Need for numerical integration
  • Deterministic integration
  • Monte Carlo integration

3. Bayesian Inference for Regression Analysis

  • Regression with non-informative prior
  • Regression with conjugate prior
  • Partially linear model
  • Regression with non-conjugate prior
  • Heteroskedastic errors
  • Autocorrelated errors
  • IID student errors

4. Bayesian Inference for vector autoregressive models

  • Unrestricted VAR and multivariate regression models
  • Posterior with NIP
  • Posterior with informative prior
  • The Minnesota prior
  • Restricted VAR and SURE models

5. Bayesian Inference for volatility models

  • ARCH models
  • Stochastic volatility models

Learning outcome knowledge

Students should be able to read critically papers and to use Bayesian inference for their own research, in each case in relation to the material that has been covered.

Exam organisation

  • Written assignment: 100%