학술논문

Necessary and Sufficient Conditions for Pareto Optimal Solution of Backward Stochastic System With Application
Document Type
Periodical
Source
IEEE Transactions on Automatic Control IEEE Trans. Automat. Contr. Automatic Control, IEEE Transactions on. 68(11):6696-6710 Nov, 2023
Subject
Signal Processing and Analysis
Pareto optimization
Games
Differential games
Costs
Optimal control
Mathematical models
Differential equations
Backward stochastic differential equation (BSDE)
closed-loop representation of open-loop optimal control
Ekeland's variational principle
Pareto cooperative differential game
portfolio and consumption selection
Language
ISSN
0018-9286
1558-2523
2334-3303
Abstract
In this article, we are interested in studying a new kind of Pareto cooperative differential game of backward stochastic differential equation. Based on the characterizations of Pareto optimal solution, the game problem is transformed into a set of single objective optimal control problems with constraints of backward stochastic differential equations. In the first place, a necessary condition for Pareto optimal strategy is established by virtue of Ekeland's variational principle, and then, it is proved that the necessary condition is also sufficient under certain convex assumption. To shed light on the application of the abovementioned theoretical results, a linear–quadratic game and a kind of optimal portfolio and consumption selection problem are also solved explicitly.