A0758
Title: Moving forward from predictive regressions: Boosting asset allocation decisions
Authors: Henri Nyberg - University of Turku (Finland) [presenting]
Lauri Nevasalmi - University of Turku (Finland)
Abstract: A flexible utility-based empirical approach is introduced to determine asset allocation decisions between risky and risk-free assets directly. This is in contrast to the commonly used two-step approach where least squares optimal statistical equity premium predictions are first constructed to form portfolio weights before economic criteria are used to evaluate resulting portfolio performance. The single-step customized gradient boosting method is designed to find optimal portfolio weights in direct utility maximization. Empirical results of the monthly U.S. data show the superiority of boosted portfolio weights over several benchmarks, generating interpretable results and profitable asset allocation decisions.
