Document Type Doctoral Thesis Author Liu, Guangling URN etd-06102008-094841 Document Title Forecasting with DSGE models : the case of South Africa Degree PhD (Economics) Department Economics Supervisor
Advisor Name Title Prof E Schaling Co-Supervisor Prof R Gupta Supervisor Keywords
- Dynamic Stochastic General Equilibrium (DSGE)
- South Africa
- Vector Autoregression Models (VARs)
Date 2008-04-16 Availability unrestricted AbstractThe objective of this thesis is to develop alternative forms of Dynamic Stochastic General Equilibrium (DSGE) models for forecasting the South African economy and, in turn, compare them with the forecasts generated by the Classical and Bayesian variants of the Vector Autoregression Models (VARs). Such a comparative analysis is aimed at developing a small-scale micro-founded framework that will help in forecasting the key macroeconomic variables of the economy.
The thesis consists of three independent papers. The first paper develops a small-scale DSGE model based on Hansen's (1985) indivisible labor Real Business Cycle (RBC) model. The results suggest that, compared to the VARs and the Bayesian VARs, the DSGE model produces large out-of-sample forecast errors.
In the basic RBC framework, business cycle fluctuations are purely driven by real technology shocks. This one-shock assumption makes the RBC models stochastically singular. In order to overcome the singularity problem in the RBC model developed in the first paper, the second paper develops a hybrid model (DSGE-VAR), in which the theoretical model is augmented with unobservable errors having a VAR representation. The model is estimated via maximum likelihood technique. The results suggest DSGE-VAR model outperforms the Classical VAR, but not the Bayesian VARs. However, it does indicate that the forecast accuracy can be improved alarmingly by using the estimated version of the DSGE model.
The third paper develops a micro-founded New-Keynesian DSGE (NKDSGE) model. The model consists of three equations, an expectational IS curve, a forward-looking version of the Phillips curve, and a Taylor-type monetary policy rule. The results indicate that, besides the usual usage for policy analysis, a small-scale NKDSGE model has a future for forecasting. The NKDSGE model outperforms both the Classical and Bayesian variants of the VARs in forecasting inflation, but not for output growth and the nominal short-term interest rate. However, the differences of the forecast errors are minor. The indicated success of the NKDSGE model for predicting inflation is important, especially in the context of South Africa - an economy targeting inflation.
© University of Pretoria 2008
Filename Size Approximate Download Time (Hours:Minutes:Seconds)
28.8 Modem 56K Modem ISDN (64 Kb) ISDN (128 Kb) Higher-speed Access 00front.pdf 105.06 Kb 00:00:29 00:00:15 00:00:13 00:00:06 < 00:00:01 01chapters1-2.pdf 265.01 Kb 00:01:13 00:00:37 00:00:33 00:00:16 00:00:01 02chapter3.pdf 234.09 Kb 00:01:05 00:00:33 00:00:29 00:00:14 00:00:01 03chapters4-5.pdf 255.89 Kb 00:01:11 00:00:36 00:00:31 00:00:15 00:00:01 04bibliography.pdf 86.89 Kb 00:00:24 00:00:12 00:00:10 00:00:05 < 00:00:01