Macroeconomic Modelling with VARs: A Stata11- Based Course

The following course will be given at the University of Technology, Sydney as five daily lectures, 10-12.30 from February 7th.

There is no charge for the course but those wishing to attend must advise Adrian Pagan by e-mail (Adrian.pagan@sydney.edu.au) by the end of January

Lecture 1 : Vector Autoregressions (VARs)

1. Introduction
1. The VAR Model
2. A Small Macro Model Example
2. Choice of model lag order (p) and model Variables
   1. Choosing p
      1. From Theoretical Models
      2. Rules of Thumb
      3. Statistical Criteria
   2. Choice of Variables
      1. Institutional Knowledge
      2. Theoretical Models
3. Restricted and Augmented VARs
   1. Restricted VARs
   2. Augmented VARs
4. Estimation and Use of a VAR
   1. Computing Impulse Responses
   2. Standard Errors of Impulse Responses

Lecture 2:  Structural VARs

1. Introduction
2. Estimating the SVAR System
   1. Recursive Systems
      1. The Small US Macro Model Again
   2. Imposing Restrictions on Impulses Responses
      1. Zero Contemporaneous Effects
         1. A Canadian Macro Model
   3. Imposing Restrictions Directly on B0 and B1
      1. A Brazilian Macro Model
3. Impulse Responses in the SVAR
4. Variance Decompositions
5. Standard Errors for Impulse Responses

Lecture 3:  Sign Restrictions on SVARs

1. Introduction
2. The VAR/SVAR Framework Again
3. Two Simple Structural Models
   1. Market ( Demand/Supply) SVAR Model
   2. A Small Macro Model
4. How do we use Sign Restriction Information?
   1. Generating a Range of Impulses
   2. Modus Operandi of Sign Restrictions
   3. An Illustration
   4. Generating Q
      1. Givens
      2. Householder Transformations
5. Identification Issues
   1. The Market Model
   2. Model Identification issues in the Macro Model
   3. How does one Deal with the Multiple Models Problem?
   4. The Multiple Shocks Problem
6. Can We Recover Correct Impulse Responses from Sign Restrictions?
7. Summary

Lecture 4: SVAR Modelling with Series that are Integrated

1. Introduction
2. SVARS with Non-cointegrated I(1) and I(0) variables
1. A Two Variable System in I(1) Variables
1. A Money Output Application
2. A Two Variable System in an I(1) and I(0) Variable
   1. 2.2.1 The Blanchard and Quah Application
3. A Multivariate System
   1. Peersman’s Early Millenium Slowdown SVAR
3. SVAR Analysis with Co-integrated I(1) an I(0) Variables
   1. Gali’s Monetary SVAR
   2. The Garratt, Lee, Pesaran, Shin U.K. Macro Model

Lecture 5: Latent Variable VARs