Gains from Switching Between Forecasts

Allan Timmermann and Yinchu Zhu

A chapter in Essays in Honor of M. Hashem Pesaran: Prediction and Macro Modeling, 2022, vol. 43A, pp 99-116 from Emerald Group Publishing Limited

Abstract: It is rare for the forecasts of one economic forecasting model to always be more accurate than the forecasts from an alternative model. This suggests the possibility of implementing a switching strategy that chooses, at each point in time, the forecasting model that is expected to be most accurate conditional on a set of instruments that are used to track the relative accuracy of the underlying forecasts. The authors analyze the factors determining the expected gains from such a switching rule over a strategy of always using one of the underlying forecasts. The authors derive bounds on the expected gains from switching for both the nested and non-nested cases and also analyze the case with a highly persistent (near-unit root) predictor variable.

Keywords: Conditional forecast accuracy; forecasting performance; real time monitoring; persistent regressor; model misspecification; time-varying parameters (search for similar items in EconPapers)
Date: 2022
References: Add references at CitEc
Citations:

Downloads: (external link)
https://www.emerald.com/insight/content/doi/10.110 ... d&utm_campaign=repec (text/html)
https://www.emerald.com/insight/content/doi/10.110 ... 1-90532021000043A006
https://www.emerald.com/insight/content/doi/10.110 ... d&utm_campaign=repec (application/pdf)
Access to full text is restricted to subscribers

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:eme:aecozz:s0731-90532021000043a006

DOI: 10.1108/S0731-90532021000043A006

Access Statistics for this chapter

More chapters in Advances in Econometrics from Emerald Group Publishing Limited
Bibliographic data for series maintained by Emerald Support ().