Construction of multi-step forecast regions of VAR processes using ordered block bootstrap
COMMUNICATIONS IN STATISTICS-SIMULATION AND COMPUTATION, vol.50, pp.2107-2125, 2021 (SCI-Expanded, Scopus)
- Publication Type: Article / Article
- Volume: 50
- Publication Date: 2021
- Doi Number: 10.1080/03610918.2019.1596282
- Journal Name: COMMUNICATIONS IN STATISTICS-SIMULATION AND COMPUTATION
- Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus, Academic Search Premier, Applied Science & Technology Source, Business Source Elite, Business Source Premier, CAB Abstracts, Compendex, Computer & Applied Sciences, Veterinary Science Database, zbMATH, Civil Engineering Abstracts
- Page Numbers: pp.2107-2125
- Keywords: Block bootstrap, Vector autoregressive model, Multivariate forecast, Forecast region
- Dokuz Eylül University Affiliated: No
Abstract
In this study, an ordered non-overlapping block bootstrap procedure has been proposed to obtain multi-step forecast regions for unrestricted vector autoregressive models. The proposed method is not based on either backward or forward representations, so it can be implemented to VARMA or VAR-GARCH models. Also, it is computationally more efficient than the existing techniques. Its finite sample performance is investigated by Monte Carlo experiments and two-real world examples. Our findings show that the proposed method is a good alternative to the available resampling methods and produces better results for long-term forecasting when the model is near non-stationary or near-cointegrated.