Factor Timing with Portfolio Characteristics

Abstract In a factor timing context, academic research has focused on identifying set of predictors that can explain the dynamics portfolios. We propose an alternative approach for portfolio returns by exploiting information from their characteristics. Different combinations dimension reduction techniques are employed to independently reduce number both and portfolios predict. Characteristic-based models outperform existing methods in terms exact predictability, as well investment performance. (JEL G10, G11, C52, C55) Received November 9, 2021; editorial decision March 23, 2023 Editor Jeffrey Pontiff. Authors have furnished Internet Appendix, which is available Oxford University Press Web site next link final published paper online.