Estimation in a Linear Model with Serially Correlated Errors When Observations are Missing
处理时间序列数据不规则间隔或观测值缺失时，带AR(1)误差的线性回归模型估计问题，推导简化变换以进行OLS估计，给出MLE及其渐近协方差矩阵，并通过真实和模拟数据比较ML与两阶段估计方法的性能。
Time series data may be obtained at irregular intervals or observations may be missing.This paper deals with the estimation of the linear regression model with AR(1) errors on the basis of such data.The etructure of the error covariance matrix is analyzed, and a simplifying transformation of the model is derived, which allows for OLS estimation.The MLE of the parameters is given and its asymptotic covariance matrix is established.The performance of ML vis-à-vis various two-stage estimation methods is assessed using both real and simulated data.