Extensions to IVX methods of inference for return predictability

The contribution of this paper is threefold. First, we demonstrate that, provided either a suitable bootstrap implementation employed or heteroskedasticity-consistent standard errors are used, the IVX-based predictability tests Kostakis et al. (2015) retain asymptotically valid inference under null hypothesis considerably weaker assumptions on innovations than required by (2015). Second, same assumptions, develop implementations IVX tests. Monte Carlo simulations show that deliver more accurate finite sample asymptotic certain problematic parameter constellations, most notably for one-sided testing, and where multiple predictors included. Third, how sub-sample approach can be used to two-sided presence temporary windows predictability.