Count data models with selectivity
展示了截断、审查、障碍、零膨胀和低报计数模型如何被解释为选择性模型，并介绍了允许计数与选择机制相关的内生选择性模型，以十年内工作变动次数为例说明方法。
This paper shows how truncated, censored, hurdle, zero inflated and underreported count models can be interpreted as models with selectivity. Until recently, users of such count data models have commonly imposed independence brtween the count generating mechanism and the selection mechanism. Such an assumption is unrealistic in most applications, and various models with endogenous selectivity (correlation between the count and the selection equations) are presented. The methods are illustrated in an application to labor mobility where the dependent variable is the number of individual job changes during a ten year period.