1. Suppose the linear regression model is如果不恒成立,则称误差项是序列相关的(自相关的)。(serial correlation) 2.产生序列相关的原因及序列相关的影响 1)原因 (1)经济行为的惯性或冲击的惯性(SARS) (2)模型误设(Model misspecification) 2、序列相关对估计与检验的影响与异方差性类似
1. Single equation estimation procedures Indirect least squares(ILS)(恰好识别) Two- -stage least squares(2SLS)(过度识别) 2. System estimation procedures Three-stage- least squares(sLS)(跨方程相 关) In1962 Theil and Zellner first proposed this method. 2SLS+GLS= 3SLS (gls=generalized least squares)