Likelihood Analysis for Probit Regression with Measurement Errors
针对一个解释变量存在测量误差且有重复测量数据的情况，提出了Probit回归模型的似然分析方法，使用EM算法计算最大似然估计，并用拉普拉斯近似进行似然比检验和置信区间计算。
Likelihood analysis is proposed for a probit regression model when one of the explanatory variables is measured with error and replicate measurements of that variable are available on some of the subjects. The distribution of the measurement error and of the unknown explanatory variable, conditional on the known variables, are both taken to be normal. The maximum likelihood estimates can be computed exactly with the EM algorithm. Likelihood ratio tests and confidence intervals can be computed with a Laplace approximation.