Beating the VAR: Improving Swedish GDP Forecasts Using Error and Intercept Corrections |
| |
| Authors: | Johan Lyhagen Stefan Ekberg Richard Eidestedt |
| |
| Affiliation: | Department of Statistics, Uppsala University, Sweden |
| |
| Abstract: | This paper examines the forecast accuracy of an unrestricted vector autoregressive (VAR) model for GDP, relative to a comparable vector error correction model (VECM) that recognizes that the data are characterized by co‐integration. In addition, an alternative forecast method, intercept correction, is considered for further comparison. Recursive out‐of‐sample forecasts are generated for both models and forecast techniques. The generated forecasts for each model are objectively evaluated by a selection of evaluation measures and equal accuracy tests. The result shows that the VECM consistently outperforms the VAR models. Further, intercept correction enhances the forecast accuracy when applied to the VECM, whereas there is no such indication when applied to the VAR model. For certain forecast horizons there is a significant difference in forecast ability between the intercept corrected VECM compared to the VAR model. Copyright © 2015 John Wiley & Sons, Ltd. |
| |
| Keywords: | forecast accuracy vector error correction vector autoregressive co‐integration intercept correction and Diebold– Mariano test |
|
|
