We develop a hybrid model that relies on the nonlinear classification Decision Tree (DT) approach but also on multivariate predictive regressions to aid implement a size rotation strategy in the U.S. equity markets. Our investment prediction is derived with a two-stage algorithm. In the first...

The paper considers time series GMM models where a subset of the parameters are time varying. We focus on an empirically relevant case with moderately large instabilities, which are well approximated by a local asymptotic embedding that does not allow the instability to be detected with...