Nonparametric Retrospection and Monitoring of Predictability of Financial Returns

We develop and evaluate sequential testing tools for a class of nonparametric tests for predictability of financial returns that include, in particular, the directional accuracy and excess profitability tests. Our sequential methods consider in a unified framework both retrospection of a historical sample and monitoring newly arriving data. To this end, we focus on linear monitoring boundaries that are continuations of horizontal lines corresponding to retrospective critical values, elaborating on both two-sided and one-sided testing. We run a simulation study and illustrate the methodology by testing for directional and mean predictability of returns in young stock markets in Eastern Europe.