UPCommons. Portal del coneixement obert de la UPC
Publicación
Contribución de congreso (ConferenceOutput)
Forecasting financial time series with multiple kernel learning
This paper introduces a forecasting procedure based on mul-tivariate dynamic kernels to re-examine –under a non linear framework–the experimental tests reported by Welch and Goyal showing that severalvariables proposed in the academic literature are of no use to predict theequity premium under linear regressions. For this approach kernel functions for time series are used with multiple kernel learning in order torepresent the relative importance of each of these variabl, Peer Reviewed