Estimation of the Duration Model by Nonparametric Maximum Likelihood, Maximum Penalized Likelihood, and Probability Simulators
用蒙特卡洛方法比较了非参数最大似然、最大惩罚似然和概率模拟器三种估计量在持续时间模型中的有限样本表现，为实证建模提供计算可行性和相对优势的参考。
Failure to properly treat heterogeneity components in longitudinal analyses can result in an incorrect parametrization of the duration model. Estimation bias is not limited to duration dependence but also extends to the structural parameters. This paper uses Monte Carlo methods to examine the finite sample behavior of three estimators for this problem: nonparametric maximum likelihood, maximum penalized likelihood, and the probability simulator. The authors' results on the estimators' finite sample behavior for this class of model add to limited experimental evidence. They highlight the estimators' computational feasibility and point to their relative strengths in empirical duration modeling. Copyright 1994 by MIT Press.