Paper: arXiv 2405.17841

Authors: Xiaomin Shi, Zuo Quan Xu

Abstract

This paper studies an optimal investment-reinsurance problem for an insurer (she) under the Cramér–Lundberg model with monotone mean–variance (MMV) criterion. At any time, the insurer can purchase reinsurance (or acquire new business) and invest in a security market consisting of a risk-free asset and multiple risky assets whose excess return rate and volatility rate are allowed to be random. The trading strategy is subject to a general convex cone constraint, encompassing no-shorting constraint as a special case. The optimal investment-reinsurance strategy and optimal value for the MMV problem are deduced by solving certain backward stochastic differential equations with jumps. In the literature, it is known that models with MMV criterion and mean–variance criterion lead to the same optimal strategy and optimal value when the wealth process is continuous. Our result shows that the conclusion remains true even if the wealth process has compensated Poisson jumps and the market coefficients are random.

Complexity vs Empirical Score

  • Math Complexity: 9.0/10
  • Empirical Rigor: 1.0/10
  • Quadrant: Lab Rats — high math complexity, limited empirical testing

Why this score: The paper heavily relies on advanced stochastic calculus, BSDEs with jumps, and random coefficient models, indicating very high mathematical density. It lacks any empirical validation, backtesting, or data implementation, focusing purely on theoretical derivation and solution existence in a controlled mathematical environment.

Research Flowchart

  flowchart TD
  A["Research Goal:<br>Optimal Investment-Reinsurance<br>under MMV Criterion"] --> B

  subgraph B ["Data & Inputs"]
      B1["Cramér--Lundberg Model<br>(Claims as Jumps)"]
      B2["Market: Risk-free &<br>Multiple Risky Assets"]
      B3["Constraints:<br>General Convex Cone"]
  end

  B --> C["Methodology:<br>Backward Stochastic Differential<br>Equations BSDEs with Jumps"]

  C --> D["Computational Process:<br>Solving BSDEs to derive<br>Optimal Strategy & Value"]

  D --> E{"Key Findings"}

  subgraph E ["Outcomes"]
      E1["Optimal Strategy<br>(Derived explicitly)"]
      E2["Optimal Value<br>(Derived explicitly)"]
      E3["Equivalence Proof:<br>MMV & Mean-Variance yield<br>identical results despite<br>jumps & randomness"]
  end

Evidence Card

What the authors claim, on what data, and who has checked it. Read from the paper's own text by a language model: every quote is verbatim, nothing was re-run. How to read this card

Proposes

The paper proposes a constrained optimal investment‑reinsurance strategy for an insurer under the Cramér–Lundberg model using a monotone mean‑variance (MMV) criterion with random market coefficients, solved via backward stochastic differential equations with jumps.

Source sentence
This paper studies an optimal investment-reinsurance problem for an insurer (she) under the Cramér--Lundberg model with monotone mean--variance (MMV) criterion. The optimal investment-reinsurance strategy and optimal value for the MMV problem are deduced by solving certain backward stochastic differential equations with jumps.

Assumes 6

  • The insurer can trade a risk‑free asset and multiple risky assets whose excess return and volatility rates are random processes. market
    Source sentence
    At any time, the insurer can purchase reinsurance (or acquire new business) and invest in a security market consisting of a risk-free asset and multiple risky assets whose excess return rate and volatility rate are allowed to be random.
  • The trading strategy must lie in a general convex cone, encompassing no‑shorting constraints. execution
    Source sentence
    The trading strategy is subject to a general convex cone constraint, encompassing no-shorting constraint as a special case.
  • Insurance claim sizes Yi are bounded i.i.d. non‑negative random variables. model
    Source sentence
    Therefore, it is reasonable to assume the claims Yi are bounded.
  • Claims arrive according to a homogeneous Poisson process with intensity λ > 0. model
    Source sentence
    As usual, we write x+ = max{x, 0} and x−= max{−x, 0} for x ∈R. 2.1 The insurance and financial model Let {Nt} be a homogeneous Poisson process with intensity λ > 0 which counts the number of claims.
Show 2 more
  • The Brownian motion W and the Poisson random measure γ are independent. statistical
    Source sentence
    3 We suppose that the Brownian motion W and the Poisson random measure γ(dt, dy) are in-dependent processes under the probability measure P. Define the filtration F = {Ft, t ⩾0} as the augmented natural filtration generated by W and γ.
  • Short‑selling of some or all stocks is prohibited in practice. execution
    Source sentence
    On the other hand, in real financial market, especially for an insure, short-selling some or all the stocks are prohibited.

Data

No dataset named in the text.

Costs

Not applicable The paper reports no trading or execution result.

Finds 2

  • The optimal investment‑reinsurance strategy and the optimal value for the MMV problem are obtained by solving certain BSDEs with jumps. theoretical
    Source sentence
    The optimal investment-reinsurance strategy and optimal value for the MMV problem are deduced by solving certain backward stochastic differential equations with jumps. Semi-closed-form solutions, the optimal probability measure and optimal investment-reinsurance in terms of the solutions to some BSDEs with jumps are provided.
  • The MMV optimal strategy and value coincide with those of the classical mean‑variance (MV) problem even when the wealth process has compensated Poisson jumps and market coefficients are random. theoretical
    Source sentence
    In the literature, it is known that models with MMV criterion and mean--variance criterion lead to the same optimal strategy and optimal value when the wealth process is continuous. Our result shows that the conclusion remains true even if the wealth process has compensated Poisson jumps and the market coefficients are random.

Checked by others

Supported by

  • Reinforcement Learning for Jump-Diffusions, with Financial Applications (2024)
    On the question: Do jumps affect the optimal mean‑variance policy and value function?
    Compare the findings

    This paper: The MMV optimal strategy and value coincide with those of the classical mean‑variance (MV) problem even when the wealth process has compensated Poisson jumps and market coefficients are random.

    That paper: In the mean‑variance portfolio selection problem, jumps do not affect the parametrization of the optimal policy or value function; both remain Gaussian/quadratic.

    Both A and B report that jumps (including compensated Poisson jumps and random coefficients) do not change the optimal mean‑variance strategy or its value, which remain as in the classical Gaussian/quadratic case.

  • A robust stochastic control problem with applications to monotone mean-variance problems (2024)
    On the question: Do optimal strategies for monotone mean‑variance and mean‑variance portfolio selection coincide in models with random coefficients?
    Compare the findings

    This paper: The MMV optimal strategy and value coincide with those of the classical mean‑variance (MV) problem even when the wealth process has compensated Poisson jumps and market coefficients are random.

    That paper: The optimal strategies for monotone mean‑variance (MMV) and mean‑variance (MV) portfolio selection problems coincide in a complete market with random coefficients.

    Both A and B report coincidence of MMV and MV optimal strategies; A adds compensated Poisson jumps, B assumes a complete market, but the result agrees.

Code

No repository by the authors found in the paper.

Extracted 2026-11-14 · same card as JSON · A language model chose these items: check the source sentence before relying on one.

Cited References (16)

  1. Introduction To Nonlinear Optimization: Theory, Algorithms, And Applications With Matlab (2017)
  2. Optimization by Vector Space Methods (1970) — link
  3. A sufficient condition for the uniform integrability of exponential martingales (1979)
  4. Semimartingale Theory and Stochastic Calculus (2019) — link
  5. Constrained Stochastic LQ Control with Random Coefficients, and Application to Portfolio Selection (2005) — link
  6. Continuous-Time Mean-Variance Portfolio Selection: A Stochastic LQ Framework (2000) — link
  7. Optimal investment–reinsurance strategy for mean–variance insurers with square-root factor process (2015) — link
  8. Filtering a Markov Modulated Random Measure (2009) — link
and 8 more
  1. PORTFOLIO SELECTION WITH MONOTONE MEAN‐VARIANCE PREFERENCES (2009) — link
  2. Continuous-Time Portfolio Choice Under Monotone Mean-Variance Preferences—Stochastic Factor Case (2019) — link
  3. Constrained stochastic LQ control with regime switching and application to portfolio selection (2022) — link
  4. A note on monotone mean–variance preferences for continuous processes (2020) — link
  5. Short Communication: Cone-Constrained Monotone Mean-Variance Portfolio Selection under Diffusion Models (2022) — link
  6. Optimal investment and reinsurance policies for the Cramér–Lundberg risk model under monotone mean-variance preference (2023) — link
  7. Constrained Monotone Mean-Variance Problem with Random Coefficients — → 2212.14188
  8. Comparison Between Mean-Variance and Monotone Mean-Variance Preferences Under Jump Diffusion and Stochastic Factor Model (2024) — link

1 of 16 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2596 linked papers.