A procedure is a specific way of making this association; a procedure is optimal if the sequence of choices it generates converges to the action that maximizes the expected payoff. The information available to the agent has crucial importance. An individual learning in isolation has partial information, because he can only observe the payoff to the action he has chosen; in social learning, he has full information, because he can learn from the action of the others. Linear procedures always con- verge to optimal action in the case of partial information, and do not in the case of full information. Exactly the opposite is true in the case of exponential procedures. Journal of Economic Literature Classification Numbers: D1, D8, D9.
Bibliographical noteFunding Information:
*The author thanks David Easley for the long discussions on a related joint work; Ramon Marimon, Harald Uhlig for discussions and advice. The comments of two referees improved the exposition considerably. He also thanks the NSF for financial support.
- Adaptive learning
- Full information
- Partial information
- Stochastic approximation