CMStatistics 2021: Start Registration
View Submission - CFE
A0206
Title: Testing the predictability of stock returns with smoothly varying deterministic mean Authors:  Matei Demetrescu - CAU Kiel (Germany) [presenting]
Mehdi Hosseinkouchak - University of Mannheim (Germany)
Abstract: Checking whether stock returns may be predicted using financial valuation ratios and other fundamental values is facing several methodological and empirical challenges. Most importantly, the typical putative predictor variable exhibits high persistence, which leads to nonstandard limiting distributions of the OLS estimator and associated t statistic in predictive regressions. While there are several methods to deal with the issue of nonstandard distributions, the high predictor persistence also opens the door to spurious regression findings induced by time-varying mean components of stock returns, if not properly controlled for. Such control requires additional information, which may not be available in practice. We take a different approach and robustify IVX predictive regression to the presence of smooth trend components. To this end, we employ a particular local mean adjustment scheme to account for possibly time-varying means. The limiting distribution of the resulting IVX t statistic is derived under sequences of local alternatives, and a wild bootstrap implementation improving the finite-sample behavior is provided. Compared to IVX predictive regression, there is a price to pay for robustness in terms of power; at the same time, the IVX statistic without adjustment consistently rejects the false null of no predictability in the presence of ignored time-varying deterministic mean components.