Title: Tests for time series momentum
Authors: Paulo Rodrigues - Universidade Noval de Lisboa (Portugal) [presenting]
Matei Demetrescu - University of Kiel (Germany)
Robert Taylor - University of Essex (United Kingdom)
Abstract: Cross-sectional momentum is a well-known asset-pricing anomaly. Consistent evidence of a new anomaly across different asset classes and markets has been found which has been termed time-series momentum. Specifically, they find that the past 12-month excess returns of an asset are a good predictor of its future. However, this conclusion is not yet consensual. For instance, standard univariate and pooled panel predictive regressions has been recently used to find little evidence of time-series momentum in the same set of assets analysed previously. These contrasting results highlight the importance of rigorous inference on the time series predictability of future returns based on past cumulative returns when testing for this phenomenon. We evaluate the performance of existing methods and propose new solutions for testing for this anomaly.