Comparing ex-ante forecasts from a sem and var model: An application to the italian economy
✍ Scribed by Gianna Boero
- Publisher
- John Wiley and Sons
- Year
- 1990
- Tongue
- English
- Weight
- 795 KB
- Volume
- 9
- Category
- Article
- ISSN
- 0277-6693
No coin nor oath required. For personal study only.
✦ Synopsis
The predictive performance of a large-scale structural econometric model (SEM) of the Italian economy-the Prometeia model-is compared in this paper with a vector autoregressive (VAR) model estimated for a selection of six main variables of interest. The paper concentrates on the quarterly ex-ante forecasts of GDP growth rate and the annual forecasts of GDP growth and inflation rate, over the period 1980-85. It concludes that no forecaster is systematically better than the other. In particular, the VAR model outperforms the SEM in short-run forecasts, suggesting that, for the latter, more careful attention should be addressed to questions of dynamic specification. On the other hand, for longer intervals, the SEM forecasts are more accurate than the VAR forecasts, in that they can benefit from the judgemental interventions of the model users and the model can pick up the non-linearities of the economy which cannot be captured by the VAR. Given the different kinds of information that can be extracted from the two approaches, it seems more reasonable to consider them as complementary rather than alternative tools for modelling and forecasting. Therefore, rather than attempting to establish the superiority of one type of model over the other, this kind of comparisons should be seen as a useful diagnostic tool for detecting types of model misspecification.
KEY WORDS Structural models Vector autoregressive models
Ex-ante forecasts Comparison of forecasts RMSEs Substantial work has been done in comparing ex-ante forecasts of large-scale structural models with time series forecasts. Most of the analysis has been addressed to UK and US models, using as a common benchmark of comparison univariate ARIMA models of the Box-Jenkins type (Holden et al., 1982; Longbottom and Holly, 1985; McNees, 1979 McNees, , 1982)). A common finding from these studies is that time series forecasts can be good competitors for SEM forecasts in the short run but not over long horizons.
More recently, time series models have moved to a multivariate context, and Bayesian vector autoregressive (BVAR) models as proposed by Litterman (1979) have been used in forecast comparisons with large-scale econometric models (see Wallis et al., 1986Wallis et al., , 1987;;Litterman, 1984; and McNees, 1986). The evidence from these studies is still very controversial. Of