A carregar...
Publicação
A deep learning test of the martingale difference hypothesis
| Nome: | Descrição: | Tamanho: | Formato: | |
|---|---|---|---|---|
| 1.21 MB | Adobe PDF |
Autores
Orientador(es)
Resumo(s)
A deep learning binary classifier is proposed to test if asset returns follow martingale difference sequences. The Neyman-Pearson classification paradigm is applied to control the type I error of the test. In Monte Carlo simulations, I find that this approach has better power properties than variance ratio and portmanteau tests against several alternative processes. I apply this procedure to a large set of exchange rate returns and find that it detects several potential deviations from the martingale difference hypothesis that the conventional statistical tests fail to capture.
Descrição
Palavras-chave
Martingale Difference Hypothesis Convolutional Network Variance Ratio Test Portmanteau Test; Exchange Rates
Contexto Educativo
Citação
Bastos João A. .(2025). “A deep learning test of the martingale difference hypothesis”. REM Working paper series, nº 0374/2025
Editora
ISEG - REM (Research in Economics and Mathematics)