Learning from peers in signaling game experiments

We investigate peer group effects in laboratory experiments based on Milgrom and Roberts' (1982, Econometrica 50: 443-459) entry limit pricing game. We generalize Heckman's (1981, in Structural Analysis of Discrete Data with Econometric Applications. MIT Press: Cambridge, MA) dynamic discrete-choice panel data models by introducing time-lagged social interactions, using the unbiased GHK simulator to implement the computationally cumbersome maximum likelihood estimation. We find that subjects' decisions are significantly influenced by past decisions of peers on several dimensions, including potential entrants' choices and strategic play of like-type monopolists. The proposed model and estimation method may be applicable to other experiments where peer group effects are likely to play an important role.