Equity allocation and portfolio selection in insurance (Q1584584)

The starting point is a Cramèr-Lundberg like utility function whose value at \(t\) is the difference between the accumulated net incomes and the accumulated claims in the time interval \([0,t]\) [cf. \textit{P. Embrechts, C. Klüppelberg} and \textit{T. Mikosch}, Modelling Extremal Events, Application of Mathematics, Springer, Berlin (1997; Zbl 0873.62116)]. The following problem is considered: allocate equity to the different portfolios (or subsidiaries), and select the portfolios, such that the annual ROEs are satisfactory, such that the probability of ruin [of a reinsurance company -- B. Ch.] and the probability of non-solvency of each portfolio is acceptable and such that the expected final overall profitability is optimal. (cf. the Introduction to the paper). It leads to an optimization problem with constraints for an unknown stochastic process (Nonlinear Model), which is an extension of the stochastic model first considered in the author's report [Allocation de fonds propres et pillotage de portefeuille. Rapport AXA-Ré et UAP Direction Scientifique, Paris, 1997]. A method of finding an approximate solution of the problem (Quadratic Model), is developed. For this model there is proved, under mild assumptions on the result processes of the portfolios, the existence of solutions in a Hilbert space of adapted (to the claim processes) square integrable processes, with the aid of the Lagrangian formalism with multipliers. The author also points out that, although the portfolios are unique, the initial equity allocation and the future dividends generically are non-unique for optimal solutions. Thus a basic arbitrage pronciple must be added to the model, in order to eliminate a degeneracy of the allocation problem. For computer simulations, in simplest cases, the reader is referred to \textit{T. Dionysopoulos} [Stochastic optimization of a simplified reinsurance portfolio. Memoir de DEA 1999, Probabilités et Applications. Université Pierre et Marie Curie, Paris (1999)].