We compare the finite sample power of short- and long-horizon tests in nonlinear predictive regression models of regime switching between bull and bear markets, allowing for time varying transition probabilities. As a point of reference, we also provide a similar comparison in a linear predictive regression model without regime switching. Overall, our results do not support the contention of higher power in longer horizon tests in either the linear or nonlinear regime switching models. Nonetheless, it is possible that other plausible nonlinear models provide stronger justification for long-horizon tests.
This paper was prepared for the 2013 Advances in Econometrics Conference: Essays in Honor of Peter C. B. Phillips held at the Southern Methodist University. We thank the conference participants, participants at the 2014 Meeting of the Canadian Economic Association, Erik Hjalmarsson, and Leo Michelis for useful discussion. We are grateful to the editors, Joon Park and Yoosoon Chang, and to an anonymous referee for very helpful comments and suggestions that have substantially improved the paper. We gratefully acknowledge the use of code provided by Erik Hjalmarsson and code made available online by Zhuanxin Ding and Marcelo Perlin. Maynard thanks the SSHRC for research funding.
Maynard, A. and Ren, D. (2014), "Assessing the Power of Long-Horizon Predictive Tests in Models of Bull and Bear Markets", Essays in Honor of Peter C. B. Phillips (Advances in Econometrics, Vol. 33), Emerald Group Publishing Limited, Bingley, pp. 673-711. https://doi.org/10.1108/S0731-905320140000033019
Emerald Group Publishing Limited
Copyright © 2014 Emerald Group Publishing Limited