Exact solution of the ruin problem in the Cramér–Lundberg model with proportional investment
Platon Promyslov, Maxim Romanov, Goluba Yurieva · 2026
Read Exact solution of the ruin problem in the Cramér–Lundberg model with proportional investment by Platon Promyslov, Maxim Romanov, Goluba Yurieva (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2604-08745.txt.
Introduction
Promyslov, Romanov, and Yurieva solve the Cramér-Lundberg ruin problem with claims and investment by share exactly. The survival chance equation reduces to a doubly confluent Heun equation, giving an exact solution in Heun functions. A verification theorem confirms the solution, and a broad analysis traces how ruin risk moves with the share invested. The short paper gives the exact solution, its proof, and its reading in terms of the share invested.
Catalog record
| CREATOR | Platon Promyslov, Maxim Romanov, Goluba Yurieva |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 32747 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2604.08745
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Exact solution of the ruin problem in the Cramér–Lundberg model with proportional investment Platon Promyslov, Maxim Romanov, Goluba Yurieva Abstract This paper considers the Cramér–Lundberg risk model with exponential claims and investments in a financial market consisting of a risk-free and a risky asset. We show that the integro-differential equation for the survival probability reduces to a linear second-order ordinary differential equation belonging to the class of doubly confluent Heun equations. The main result of the paper is the construction of an exact analytical solution to the problem, expressed in terms of Heun functions, and the proof of a verification theorem. The obtained formula allows for an exact quantitative analysis for any initial capital, revealing a nontrivial dependence of the company’s reliability on the chosen investment strategy. ∎ MSC 91G05 60G51 45J05 1 Introduction The classical collective risk theory, dating back to the works of Lundberg and Cramér, describes the capital reserve of an insurance company as a process with a constant deterministic drift perturbed by a Poisson stream of claims. In modern actuarial mathematics, this paradigm is extended: it is assumed that the company’s capital is not only accumulated but also continuously invested in a financial market. The inclusion of a risky asset, the price of which follows a geometric Brownian motion, fundamentally changes the dynamics of the system: even with light (e.g., exponential) tails of the claim size distribution, the ruin probability acquires a heavy power-law asymptotic behavior Frolova2002 ; Kalashnikov2002 ; Paulsen1993 . Two main analytical approaches to studying the ruin probability in models with investments have emerged in the literature. The first one is based on the implicit renewal theory Goldie1991 . This method allows obtaining exact decay rates of the ruin probability at infinity for a broad class of processes (see, e.g., Albrecher2012 ; Eberlein2022 ; KLP2025 ; KPro2023 ; Pergamenshchikov2006 ). However, being asymptotic by its nature, this approach fundamentally provides no information about the behavior of the survival function for finite values of the initial capital. The second approach is based on the analysis of integro-differential equations, which, by virtue of martingale properties and the generalized Itô formula, are satisfied by the survival probability. Within this direction, questions of smoothness, existence, and uniqueness of classical or viscosity solutions are investigated. In Grandits2004 , the existence of a weak solution, i.e. W_{\mathrm{loc}}^{2,1} , was proved under the assumptions that F is continuous in a neighbourhood of zero and that \mathbb{E}[\xi^{\varepsilon}]<\infty for some \varepsilon>0 . Moreover, a Karamata-type theorem was established, connecting the asymptotic behavior of the ruin probability with the regular variation of the tail of the claim size distribution. In AK2026 , the existence of a classical solution required the moment condition \mathbb{E}[\xi^{\gamma-1}]<\infty. (1) Recently, in P2026 , it was shown that for the classical Cramér–Lundberg model with proportional investment, the existence of a C^{2} -solution also does not require the moment \gamma-1 ; it suffices to have an arbitrarily small moment \mathbb{E}[\xi^{\varepsilon}]<\infty and continuity of the distribution function. Condition ( 1 ) is needed only to obtain the exact power-law asymptotics of the ruin probability. Despite a deep understanding of the qualitative properties of these equations, explicit closed-form analytical solutions are practically absent in the literature. The lack of exact formulas forces researchers to rely on asymptotic estimates or numerical simulations, which inevitably leads to significant errors for small and medium values of the initial capital—precisely the region of greatest practical interest. The present paper aims to fill this gap. The main objective of this study is to construct an exact analytical solution to the ruin problem in the Cramér–Lundberg model with proportional investment and exponentially distributed claims. Our methodology relies on reducing the IDE to a second-order linear ordinary differential equation (ODE) with respect to the derivative of the survival probability. The key observation is that, after a suitable conformal change of variables, the resulting ODE is strictly classified as a doubly confluent Heun equation—an equation with two irregular singular points on the extended complex plane. This allows us, for the first time in the context of actuarial mathematics, to express the survival probability density in terms of special Heun functions. The main result of the paper is an explicit analytical formula for the survival probability, valid for any initial capital u\geq 0 . We perform an asymptotic analysis of the fundamental systems of solutions in the neighborhoods of the singularities, isolate the uniquely physically meaningful (integrable) solution, and prove a verification theorem guaranteeing that the constructed function is identical to the survival probability of the stochastic process under consideration. The availability of an exact solution makes it possible not only to confirm known asymptotic results but also to investigate in detail the non-monotonic impact of the share of risky investments on the company’s reliability profile. The paper is organized as follows. Section 2 formalizes the stochastic model of capital dynamics and states the mathematical problem. In Section 3 , the IDE is reduced to a differential equation for the density. Section 4 is devoted to the analysis of the singularities of the obtained ODE and its reduction to the canonical form of the Heun equation. In Section 5 , the fundamental systems of solutions are constructed, and the exact formula for the survival probability is derived. Section 6 provides the proof of the verification theorem based on the theory of local martingales. The concluding Section 7 contains numerical illustrations and a qualitative analysis of the constructed solution. 2 Model and problem formulation Let a stochastic basis (\Omega,\mathcal{F},\mathbf{F},\mathbf{P}) be given, endowed with a filtration \mathbf{F}=(\mathcal{F}_{t})_{t\geq 0} satisfying the usual conditions (right-continuity and completeness with respect to \mathbf{P} -null sets). On this space, the following independent stochastic objects are defined: a one-dimensional \mathbf{F} -adapted standard Brownian motion W=(W_{t})_{t\geq 0} ; a homogeneous Poisson process N=(N_{t})_{t\geq 0} with intensity \lambda>0 ; and a sequence \xi=(\xi_{k})_{k\geq 1} of independent and identically distributed random variables, independent of W and N . In the framework of this paper, we assume that the variables \xi_{k} follow an exponential distribution with parameter \mu>0 and cumulative distribution function F(x)=1-e^{-\mu x} for x\geq 0 . The aggregate claim process is a compound Poisson process given by Z_{t}=\sum_{k=1}^{N_{t}}\xi_{k}. Consider an insurance company with an initial capital u\geq 0 . The premium income is collected deterministically at a constant rate c>0 . The company’s capital is continuously invested: a fixed proportion \kappa\in(0,1] is allocated to a risky asset, while the remaining part 1-\kappa is kept in a risk-free bank account. The asset price dynamics are governed by the equations: dB_{t}=rB_{t}\,dt,\qquad dS_{t}=S_{t}\bigl(a\,dt+\sigma\,dW_{t}\bigr), where r\geq 0 is the risk-free interest rate, a>r is the expected return of the risky asset, and \sigma>0 is its volatility. The capital process X=(X_{t})_{t\geq 0} with the initial condition X_{0}=u satisfies the following stochastic differential equation (SDE) with jumps: dX_{t}=(c+a_{\kappa}X_{t})\,dt+\sigma_{\kappa}X_{t}\,dW_{t}-dZ_{t},\quad X_{0}=u, (2) where we have introduced the effective parameters of the investment portfolio: a_{\kappa}:=r+\kappa(a-r),\qquad\sigma_{\kappa}:=\kappa\sigma. Since we exclude the degenerate case \kappa=0 , the diffusion component in ( 2 ) is strictly positive for X_{t}>0 . The time of ruin \tau is defined as the first exit time of the capital process into the strictly negative half-line: \tau=\inf\{t>0:X_{t}<0\}, with the convention \inf\varnothing=\infty . Because the diffusion coefficient vanishes at X_{t}=0