Dynamic R&D Competition with Learning
构建了一个未知风险率的研发竞争微分博弈模型，发现随着时间推移，企业会因成功而变得悲观，减少研发投入甚至退出；竞争企业数量增加会提高研发强度，但信念演化速度不同导致研发路径并非单调排序。
To account for the possibility that firms are unsure about the ease of innovation, we formulate a differential game of R&D competition with an unknown hazard rate. We show, as time passes with success, firms become more pessimistic about eventual innovation, reducing their R&D investment and possibly exiting the race. An increase in the number of competing firms tends to increase firms' R&D intensities, for given beliefs, but because beliefs evolve at different rates depending on the number of firms in the race, time paths of R&D investment intensity are not unambiguously ordered with respect to the number of competing firms.