Endogenous treatment effects for count data models with endogenous participation or sample selection

Department of Quantitative Social Science Endogenous Treatment Effects for Count Data Models with Sample Selection or Endogenous Participation

In this paper we propose a method to estimate models in which an endogenous dichotomous treatment affects a count outcome in the presence of either sample selection or endogenous participation using maximum simulated likelihood. We allow for the treatment to have an effect on both the sample selection or the participation rule and the main outcome. Applications of this model are frequent in man...

A Bayesian Analysis of Endogenous Switching Models for Count Data

Censored Regression, Sample Selection, Endogenous Switching, and Treatment-Effect Regression Models

Truncated and censored regression models are useful when the dependent variable is observed in some ranges but not in others. A classical example is Tobin’s (1958) study of household expenditures, in which expenditures on durable goods were seen to generally increase with income, but a non-trivial number of households had no expenditures on durable goods. The problem, in this case, is how to mo...

Committee design with endogenous participation

We investigate the optimal design of a committee in a model with the endogenous participation of experts who have private information about their own abilities. We study three different dimensions of committee design: members’ wages, the number of seats, and the communication system. We show that, surprisingly, higher wages lead to lower quality experts. By contrast, transparency improves the q...

Properties of estimators of count data model with endogenous switching

We examine properties of estimators of count data model with endogenous switching. The estimation of the count data model that accommodates endogenous switching can be accomplished by Full Information Maximum Likelihood (FIML). However, FIML estimation requires fully and correctly specified model and is computationally burdensome. Alternative estimation methods do not require fully specified mo...