LINEAR SYSTEMS WITH MARKOV JUMPS AND MULTIPLICATIVE NOISES - CONSTRAINED TOTAL VARIANCE PROBLEM.
Fabio Barbieri1; Oswaldo L. V. Costa1
1 Universidade de São Paulo
Resumo
In this paper we consider the stochastic multi-period optimal control problem of discrete-time linear systems subject to Markov jumps and multiplicative noises under a new criteria. We develop a solution for a control policy that maximizes the expected return while keeping the total variance weighted by its risk aversion parameter lower than some maximum value. We conclude the paper by presenting a numerical example of a portfolio selection problem.
Palavras-chave: Mean-Variance Control; Markov Jumps Systems; Multiplicative Noises; Stochastic Optimal Control; Riccati Equations; Constrained Variance