质量随机的研发竞赛模型与最优奖励

假设研发质量具有随机性，建立了研发竞赛的非合作博弈模型，分别探讨了完全信息和不完全信息条件下竞赛参与人与竞赛发起者的最优策略。研究发现：竞赛参与人的研发投入水平在完全信息和不完全信息条件下都随自身研发效率的降低而降低；研发参与人的类型信息不完全程度越高，参与人的均衡研发投入越低；在每一种信息类型组合下都存在着最优奖励使得竞赛发起者的期望收益最大。

R&D Race Model with Stochastic Quality and Optimal Incentive

Based on the ty of R＆D, the paper builds of R＆D contest, discusses assumption of stochastic qualia non-cooperative game model the optimal strategies of the sponsor and the contestants under complete information and incomplete information. The results show that the input lev- el of the contestants both decline with their R＆D efficiency under complete and incomplete information, input level of the contestants declines with the increase of incompleteness degree of the information on the type of the contestants, and there exists an optima fixed prize for the sponsor gaining maximum expected revenue under every combination of contestants types.