Testing for equal predictive accuracy with strong dependence

IF 6.9 2区经济学 Q1 ECONOMICS

International Journal of Forecasting Pub Date : 2024-12-05 DOI:10.1016/j.ijforecast.2024.11.003

Laura Coroneo , Fabrizio Iacone

引用次数: 0

Abstract

We analyse the properties of the Diebold and Mariano (1995) test in the presence of autocorrelation in the loss differential. We show that the power of the Diebold and Mariano (1995) test decreases as the dependence increases, making it more difficult to obtain statistically significant evidence of superior predictive ability against less accurate benchmarks. We also find that, after a certain threshold, the test has no power, and the correct null hypothesis is spuriously rejected. These results caution us to seriously consider the loss differential’s dependence properties before applying the Diebold and Mariano (1995) test.

查看原文本刊更多论文

具有强依赖性的相等预测精度的测试

我们分析了Diebold和Mariano（1995）在损失微分中存在自相关的情况下检验的性质。我们表明，Diebold和Mariano（1995）检验的效力随着依赖性的增加而降低，这使得在较不准确的基准上获得具有优越预测能力的统计显著证据变得更加困难。我们还发现，在某个阈值之后，检验没有效力，正确的零假设被虚假地拒绝。这些结果提醒我们在应用Diebold和Mariano（1995）检验之前要认真考虑损失微分的依赖性质。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

International Journal of Forecasting Multiple-

CiteScore

17.10

自引率

11.40%

发文量

189

审稿时长

77 days

期刊介绍： The International Journal of Forecasting is a leading journal in its field that publishes high quality refereed papers. It aims to bridge the gap between theory and practice, making forecasting useful and relevant for decision and policy makers. The journal places strong emphasis on empirical studies, evaluation activities, implementation research, and improving the practice of forecasting. It welcomes various points of view and encourages debate to find solutions to field-related problems. The journal is the official publication of the International Institute of Forecasters (IIF) and is indexed in Sociological Abstracts, Journal of Economic Literature, Statistical Theory and Method Abstracts, INSPEC, Current Contents, UMI Data Courier, RePEc, Academic Journal Guide, CIS, IAOR, and Social Sciences Citation Index.