Using Subjective Expectations Data to Allow for Unobserved Heterogeneity in Hotz-Miller Estimation Strategies

Juan Pantano, Yu Zheng
2010 Social Science Research Network  
We introduce a novel approach to allow for unobserved heterogeneity in two-step structural estimation strategies for discrete choice dynamic programming models (i.e strategies that avoid full solution methods). We contribute to the literature by adopting a ...xed e¤ ects approach: rather than identifying an unobserved heterogeneity distribution, we actually reveal the true unobserved type of each observation in a ...rst step. We do so by exploiting the tight link between the conditional choice
more » ... conditional choice probabilities that are derived from the economic model and just two subjective self-reported assessments about future choice probabilities such as those commonly elicited in major surveys. We uncover the unusual power of ideal expectations data to identify unobserved types for di¤erent classes of models. Of more empirical relevance, we show that our results hold when we allow these subjective future choice probabilities to be elicited in less than ideal circumstances, such as, for example, when self-reports display substantial "heaping" at "focal" reference values. tural parameter vector in the estimation routine. 1 The seminal work of Hotz & Miller (1993) shows how to estimate the structural parameters of a discrete choice dynamic programming model without solving the optimization problem even once. The Hotz-Miller strategy has generated abundant work on estimation of structural models that builds upon this initial insight 2 . However, an inherent problem in the Hotz-Miller type of strategy exploited by these papers is that, because of its very own nature, it cannot accommodate permanent sources of unobserved heterogeneity: The ...rst step recovers equilibrium behavior policies from the data, and as such, these can only be recovered based on observables. On the other hand, the more computationally intensive "frontal strategies" can handle permanent unobserved heterogeneity by integrating out the unobserved types in the likelihood function. 3
doi:10.2139/ssrn.2129303 fatcat:7wt62qrx4ngvldbh7kzhifk7tq