Optimal Stochastic Robust Controllers based on Mean‎- ‎CVaR Sliding Function and its Application in Pairs Trading Strategy

Articles in Press

Document Type : Research Paper

Authors

1 Department of Mathematics and Computer Science, Lorestan University, Khorramabad, Lorestan 44316- 68151, Iran.

2 1. Department of Mathematics and Computer Science, Lorestan University, Khorramabad, Lorestan 44316- 68151, Iran. 2. Faculty of Mathematics and Statistics, Isfahan University, Isfahan, 03137934611, Iran.

Abstract

‎Robust control of some nonlinear stochastic risky systems‎, ‎in the presence of uncertainties and undesirable risks‎, ‎requires the use of suitable and coherent risk measurement tool and efficient filters‎. ‎In this paper‎, ‎two new optimal robust stochastic controllers are designed for nonlinear stochastic risk systems and applications in financial engineering‎. ‎For this purpose‎, ‎a new weighted conditional sliding function is designed based on a new convex‎, ‎coherent and differentiable risk measure of the state variables‎. ‎Also‎, ‎a new efficient frontier with slope CVaR-Sharp‎- ‎ratio is designed to optimally estimate the weight of the CVaR measure in the proposed conditional sliding function‎, ‎based on the efficient frontier analysis approach‎. ‎Additionally‎, ‎the adaptive boundary layer width and robust controllers are tuned to increase the performance of the proposed controllers‎. ‎Also‎, ‎four theorems are proved and a new algorithm is designed for stability analysis and computational support‎. ‎Finally‎, ‎to demonstrate the advantages of the proposed methods‎, ‎a pair trading stock problem is simulated‎. The simulation results show that the proposed methods improve portfolio returns and achieve effective CVaR-based risk control.

Keywords

Main Subjects

Computational Methods for Differential Equations

Articles in Press, Accepted Manuscript
Available Online from 26 July 2026

Files

History

  • Receive Date: 14 August 2025
  • Revise Date: 10 July 2026
  • Accept Date: 12 July 2026

Share

How to cite

Statistics