Premium
Design of model predictive control for constrained Markov jump linear systems with multiplicative noises and online portfolio selection
Author(s) -
Dombrovskii Vladimir,
Pashinskaya Tatiana
Publication year - 2019
Publication title -
international journal of robust and nonlinear control
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.361
H-Index - 106
eISSN - 1099-1239
pISSN - 1049-8923
DOI - 10.1002/rnc.4807
Subject(s) - mathematical optimization , multiplicative function , model predictive control , state vector , computer science , markov chain , portfolio , mathematics , control theory (sociology) , control (management) , artificial intelligence , machine learning , finance , mathematical analysis , physics , classical mechanics , economics
Summary In this paper, we consider model predictive control for a class of constrained discrete‐time Markov jump linear systems with multiplicative noises. A generalized performance criterion is composed of a weighted sum of a linear combination of the (a) expected value of quadratic forms of state and control vectors, (b) quadratic forms of the expected value of the state vector, and (c) the linear component of the expected value of the state vector. The goal of the present paper is to design optimal control strategies subject to hard constraints on the input manipulated variables and to provide a numerically tractable algorithm for practical applications. The results are applied to a problem of online investment portfolio selection. Our approach is tested on a set of a real data from the New York Stock Exchange.