Combining p -values for Multivariate Predictive Ability Testing
Bibliographic Data
| ID | 19417537 |
|---|---|
| Authors | Lars Spreng (Centre for Econometric Analysis and Faculty of Finance Bayes Business School (formerly Cass), London, UK, corresponding author), Giovanni Urga (0000-0002-6742-7370, Centre for Econometric Analysis and Faculty of Finance Bayes Business School (formerly Cass), London, UK) |
| Year | 2023 |
| Volume | 41 |
| Issue | 3 |
| Pages | 765-777 |
| Publication date | 2023-07-03 |
| Open Access | Yes |
| Type | ARTICLE |
| Venue | Journal of Business and Economic Statistics (JOURNAL) |
| Journal identifiers | ISSN: 0735-0015 • E-ISSN: 1537-2707 |
| Publisher | Informa UK Limited (PUBLISHER • GB) |
| DOI | 10.1080/07350015.2022.2067545 |
| OpenAlex | W4224272103 |
| Language | EN |
| References cited | 16 |
In this article, we propose an intersection-union test for multivariate forecast accuracy based on the combination of a sequence of univariate tests. The testing framework evaluates a global null hypothesis of equal predictive ability using any number of univariate forecast accuracy tests under arbitrary dependence structures, without specifying the underlying multivariate distribution. An extensive Monte Carlo simulation exercise shows that our proposed test has very good size and power properties under several relevant scenarios, and performs well in both low- and high-dimensional settings. We illustrate the empirical validity of our testing procedure using a large dataset of 84 daily exchange rates running from January 1, 2011 to April 1, 2021. We show that our proposed test addresses inconclusive results that often arise in practice
Data mining · Econometrics · Monte Carlo method · Multivariate analysis · Multivariate statistics · Null hypothesis · Predictive power · Statistics · Univariate · Computer Science · Engineering · Financial Risk and Volatility Modeling · Forecasting Techniques and Applications · Mathematics · Monetary Policy and Economic Impact
| Highly cited | No |
|---|