Semiparametric censored regression models
Semiparametric censored regression models
复制标题
DOI：
10.1257/jep.15.4.29
复制
发表时间：
2001-09-01
影响因子：
8.4
通讯作者：
Powell, JL
中科院分区：
文献类型：
--
作者：
Chay, KY;Powell, JL
当数据被删除时，普通的最小二乘回归可以提供有偏的系数估计。这个问题的最大似然方法只有在正确指定误差分布的情况下才有效，这在实践中可能是有问题的。我们回顾了删失回归模型的几种半参数估计，它们不需要对误差分布进行参数化。这些估计值被用来检查20世纪60年代基于审查的税收记录的黑人和白人收入不平等的变化。结果表明，1964年民权法案通过后，美国南部黑人和白人男性的收入出现了显著的趋同。
When data are censored, ordinary least squares regression can provide biased coefficient estimates. Maximum likelihood approaches to this problem are valid only if the error distribution is correctly specified, which can be problematic in practice. We review several semiparametric estimators for the censored regression model that do not require parameterization of the error distribution. These estimators are used to examine changes in black-white earnings inequality during the 1960s based on censored tax records. The results show that there was significant earnings convergence among black and white men in the American South after the passage of the 1964 Civil Rights Act.