Are short term stock asset returns predictable? : An extended empirical analysis
In this paper, the question is addressed, whether or not short term stock asset returns are predictable from the knowledge of the past return series. Several methods like Hodrick-Prescott-filtering, Kalmanfiltering, GARCH-M, EWMA and non-parametric regression are benchmarked against naive mean predictions. The extended empirical analysis includes data of several hundred stocks and indexes over the last ten years. The root mean square error of the concerning prediction methods is calculated for several rolling windows of different length, in order to assess whether or not local model fit is favorable.
Nutzung und Vervielfältigung:
Alle Rechte vorbehalten