该文考虑协变量缺失时的多元线性EV模型参数的估计, 其中协变量的缺失机制是Rubin(1976)提出的随机缺失(MAR). 利用加权调整最小二乘方法给出参数估计, 证明了估计的相合性和渐近正态性. 数值模拟结果表明所给的估计性态良好.
The multivariate linear errors-in-variables model when the regressors are missing at random in the sense of Rubin (1976) is considered in this paper. The weighted adjust LS parameter estimation based on inverse probability is proposed. It is shown that the estimators have consistency and asymptotic normality.Simulations are shown that our estimators perform well.