Belmont University Research Symposium (BURS)
Comparing Predictive Accuracy of Multiple Forecasts
Publication Date
4-2024
College
Sciences and Mathematics, College of
Department
Math and Computer Science, Department of
BURS Faculty Advisor
Dr. William Best
Presentation Type
Poster Presentation
Abstract
We investigate a modification of the Mariano-Preve (2012) Chi Square test for equal predictive ability of three or more forecasting models. Under the proposed modification, the null distribution of the Wald-type test statistic is asymptotically Hotelling T^2. The modified version of the test is empirically shown to outperform the Chi-Square version. Monte Carlo simulations indicate improved empirical size for series of moderate lengths. For series of small length the empirical size is larger than nominal, but is an improvement over the Chi Square. A finite sample correction factor is moderately successful in correcting the size of the test. Finally, an investigation into the empirical size for various significance levels found an increased error rate for smaller significance levels, suggesting asymptotic issues which are the topics of future research.
Recommended Citation
Farnsworth, Matt and Best, Will, "Comparing Predictive Accuracy of Multiple Forecasts" (2024). Belmont University Research Symposium (BURS). 395.
https://repository.belmont.edu/burs/395
