In this new paper, João Nicolau and colleagues introduce a portfolio optimization procedure that minimizes intra-horizon risk.
The authors use a nonparametric method for estimating first hitting time probabilities, and a novel approach to Markov chain order selection. This optimization framework proposed thus allows to include novel path-dependent measures of risk and return in the asset allocation problem.
The authors even provide an empirical application to S&P 100 companies, a risk-free asset and stock indices. These empirical results suggest that the proposed framework indeed exhibits more consistency and an alternative optimization problem
