Estimation of sample selection bias models
研究了最大似然估计（MLE）在样本选择偏差模型中的计算方法，并通过蒙特卡洛实验和实证例子比较了MLE与Heckman两步估计的有限样本性质。
Econometric models with sample selection biases are widely used in various fields of economics, such as labor economics. The Maximum Likelihood Estimator (MLE) is seldom used to estimate models because of computational difficulty, while Heckman's two-step estimator is widely used to estimate these models. However, Heckman's two-step estimator sometimes performs poorly. In this paper, methods of calculating the MLE are analysed, and finite sample properties of the MLE and Heckman's two-step estimator are compared using Monte Carlo experiments and empirical examples.