Verification Theory and Approximate Optimal Harvesting Strategy for a Stochastic Competitive Ecosystem Subject to Lévy Noise

Abstract

This work focuses on optimal harvesting problems for a stochastic competitive ecosystem subject to Levy noise. A verification theorem for corresponding harvesting strategy is established, which offers sufficient conditions for deriving an optimal harvesting strategy and an upper bound of the value function. For a given instantaneous marginal yields function, a concrete upper bound of value function is constructed by applying the verification theorem obtained in this paper. Meanwhile, the monotonicity of value function is investigated. Also, an e-optimal harvesting strategy is designed to find an approximate optimal harvesting strategy for those harvesting problems with no exact optimal harvesting strategy. Finally, by choosing appropriately Markov decision process defined on a discrete state space, a computational method for an optimal harvesting strategy is designed and a concrete example is also given to show the implementation of the algorithm.