Relaxing the symmetry assumption in participation games: a specification test for cluster-heterogeneity

Abstract We propose a novel approach to check whether individual behaviour in binary-choice participation games is consistent with the restrictions imposed by symmetric models. This allows particular an assessment of how much cluster-heterogeneity model can tolerate remain its behavioural restrictions. assess our data from market-entry experiments which we analyse through lens ‘Exploration versus Exploration’ (EvE, equivalent Logit-QRE) or Impulse Balance Equilibrium (IBE). find that when symmetry assumption imposed, both models are typically rejected assuming pooled and IBE yields more data-consistent estimates than EvE, i.e., IBE’s session those EvE. When relaxing symmetry, EvE (IBE) for 17% (42%) time. Although support cluster-heterogeneity, less likely yield over-parametrised specifications insignificant so it outperforms accommodating model-consistent cluster-heterogeneity. The use regularisation procedures estimations partially addresses EvE’s shortcomings but leaves overall conclusions unchanged.