loading page

Deep learning solution of optimal reinsurance-investment strategies with extra information and multiple risks
  • Fanyi Peng,
  • Ming Yan,
  • Shuhua Zhang
Fanyi Peng
Tianjin University of Finance and Economics
Author Profile
Ming Yan
Tianjin University of Finance and Economics
Author Profile
Shuhua Zhang
Tianjin University of Finance and Economics

Corresponding Author:[email protected]

Author Profile

Abstract

This paper investigates an optimal investment-reinsurance problem for an insurer who possesses extra information regarding the future realizations of the claim process and risky asset process. The insurer sells insurance contracts, has access to proportional reinsurance business, and invests in a financial market consisting of three assets: one risk-free asset, one bond and one stock. Here, the nominal interest rate is characterized by the Vasicek model; and the stock price is driven by the Heston’s stochastic volatility model. Applying the enlargement of filtration techniques, we establish the optimal control problem in which an insurer maximizes the expected power utility of the terminal wealth. By using the dynamic programming principle, the problem can be changed to four-dimensional Hamilton-Jacobi-Bellman equation. In addition, we adopt a deep neural network method by which the partial differential equation is converted to two backward stochastic differential equations and solved by a stochastic gradient descent-type optimization procedure. Numerical results obtained using TensorFlow in Python and the economic behavior of the approximate optimal strategy and the approximate optimal utility of the insurer are analyzed.
17 Oct 2023Submitted to Mathematical Methods in the Applied Sciences
18 Oct 2023Submission Checks Completed
18 Oct 2023Assigned to Editor
26 Oct 2023Review(s) Completed, Editorial Evaluation Pending
28 Oct 2023Reviewer(s) Assigned
20 Mar 2024Editorial Decision: Revise Major
31 May 20241st Revision Received
05 Jun 2024Submission Checks Completed
05 Jun 2024Assigned to Editor
05 Jun 2024Review(s) Completed, Editorial Evaluation Pending
20 Aug 2024Editorial Decision: Accept