This paper compares the performance of stimulus response (SR) and belief-based learning (BBL) using data from game theory experiments. The environment, extensive form games played in a population setting, is novel in the empirical literature on learning in games. Both the SR and BBL models fit the data reasonably well in common interest games with history while the test results accept SR and reject BBL in games with no history and in all but one of the divergent interest games. Estimation is challenging since the likelihood function is not globally concave and the results may be subject to convergence bias.
|Place of Publication||Tilburg|
|Number of pages||48|
|Publication status||Published - 2000|
|Name||CentER Discussion Paper|
- game theory and experiments