Selecting a Nonlinear Time Series Model using Weighted Tests of Equal Forecast Accuracy
Dick van Dijk () and
Philip Hans Franses
No EI 2003-10, Econometric Institute Research Papers from Erasmus University Rotterdam, Erasmus School of Economics (ESE), Econometric Institute
Nonlinear time series models have become fashionable tools to describe and forecast a variety of economic time series. A closer look at reported empirical studies, however, reveals that these models apparently fit well in-sample, but rarely show a substantial improvement in out-of-sample forecasts, at least over linear models. One of the many possible reasons for this finding is that inappropriate model selection criteria and forecast evaluation criteria are used. In this paper we therefore propose a novel criterion, which we believe does more justice to the very nature of nonlinear models. Simulations show that our criterion outperforms currently used criteria, in the sense that the true nonlinear model is more often found to perform better in out-of-sample forecasting than a benchmark linear model. An empirical illustration for US GDP emphasizes its relevance.
Keywords: forecast evaluation; forecasting; model selection; nonlinearity (search for similar items in EconPapers)
JEL-codes: C22 C52 C53 E32 E37 (search for similar items in EconPapers)
References: View references in EconPapers View complete reference list from CitEc
Citations: View citations in EconPapers (20) Track citations by RSS feed
Downloads: (external link)
Journal Article: Selecting a Nonlinear Time Series Model using Weighted Tests of Equal Forecast Accuracy* (2003)
This item may be available elsewhere in EconPapers: Search for items with the same title.
Export reference: BibTeX
RIS (EndNote, ProCite, RefMan)
Persistent link: https://EconPapers.repec.org/RePEc:ems:eureir:1703
Access Statistics for this paper
More papers in Econometric Institute Research Papers from Erasmus University Rotterdam, Erasmus School of Economics (ESE), Econometric Institute Contact information at EDIRC.
Bibliographic data for series maintained by RePub ().