Collective learning, innovation and growth in a boundedly rational, evolutionary world

摘要: We formulate a simple multiagent evolutionary scheme as model of collective learning, i.e. situation in which firms experiment, interact, and learn from each other. This is then applied to stylized endogenous growth economy have determine how much invest R&D, where innovations are the stochastic product their R&D activity, spillovers occur, but technological advantages only relative temporary actually diffuse, both at intra interfirm levels. The demonstrates existence unique long-run attractor (in linear case) distinct phases on road that attractor. also compare patterns for logistic innovation function, produce some evidence bifurcation latter case.