The National Centre for Econometric Research (NCER) will hold a 5 (part) day course covering Dynamic Stochastic General Equilibrium modelling at the Gardens Point campus, Queensland University of Technology, Brisbane, from Monday 22 to Friday 26 September 2008. Afternoon tea will be provided on all days. If you are interested in attending, please contact the Project Officer.
Workshop venue is the Dennis Gibson Room, Level 10, Z Block, Z1064, QUT. (Please see the Gardens Point map for location information.)
Workshop Convener
Professor Adrian Pagan, Queensland University of Technology/University of New South Wales.
Course Description
The course will look at the construction and use of dynamic macro-economic models. It places these in an historical context and shows how these models have evolved out of earlier versions in order to be more “complete”.
Macro-economic models study aggregates of an economy. In the past two decades models that emphasise aggregates have moved from being static, and assuming perfect foresight, to being dynamic with an emphasis upon the substantial amount of uncertainty in aggregate economic outcomes. The latter is generally characterized by the idea that shocks occur which are not perfectly predictable.
The course begins by studying the basic Solow-Swan growth model. This model assumes a constant savings rate and is often generalized with the Ramsey model. Introducing stochastic elements into the Ramsey model produces the stochastic growth model ( also known as the Real Business Cycle model). Generally the RBC model fails to replicate the dynamics of actual economies and some extra features need to be added to it for that purpose. Mostly these are designed to account for the fact that adjustments take time. Common ways of doing this are the addition of habits to make consumption choices smoother and adjustment costs to investment to make that smoother.
For purposes of stabilization policy it is necessary to describe inflation and how nominal rigidities make variation in nominal interest rates have real effects. Some standard ways of doing this, such as Calvo pricing, are briefly described. This leads on to a simple models that incorporate these features, such as the New Keynesian “gaps” model and Dynamic Stochastic General Equilibrium Models (DSGE). These models have proven to be useful for discussing macro-economic stabilization policy. Many of the techniques used in methods to introduce nominal price rigidities derive from CGE models so providing a connection to that literature.
Extending DSGE models to open economies requires some thought and the models need to be designed to ensure that they will produce sensible outcomes. We look at how this has been done.
Finally we provide a summary of the major ways that estimation of these parameters has been performed.
Throughout the course after each lecture we will illustrate how one can work with these models using the Dynare package. Dynare can be downloaded from http://www.cepremap.cnrs.fr/dynare/. I am using version 3.065 for MATLAB. It is worth getting the Dynare User Guide by Tommaso Mancini which is available from the web page above under the documentation section.
Sessions
The first 2 hours of each session will be devoted to a lecture. Then there will be a break of 15-20 mins and the remaining time will be used for dealing with more technical issues such as applications of the Dynare program
| Day |
Time |
Topic |
| 1 |
12:00 pm to 3:00 pm |
Modelling Designs in History - Solow Swan and CGE Models |
| 2 |
12:30 pm to 3:30 pm |
The Stochastic Growth (RBC) Model |
| 3 |
12:00 pm to 3:00 pm |
Introducing Nominal and Real Rigidities- New Keynesian and DSGE Models |
| 4 |
12:00 pm to 3:00 pm |
Open Economy Models |
| 5 |
01:00 pm to 4:00 pm |
Estimation of Model Parameters |
Enquiries
For further details please contact the Project Officer:
Angela Fletcher
Queensland University of Technology
Email: a.fletcher@qut.edu.au
