Chapter 12 联立方程模型的估计
1,Single equation estimation procedures
1) Indirect least squares (ILS) (恰好识别 );
2) Two-stage least squares (2SLS) (过度 识别 )
2,System estimation procedures
Three-stage least squares (3SLS) (跨方程相关 )
In 1962,Theil and Zellner first proposed this
method.
2SLS+GLS= 3SLS
( GLS=generalized least squares )
Example (Klein’s Model)
Consumption:
CT=C(1)+C(2)*P+C(3)*P(-1)+C(4)*(W1+W2)
Investment:
I=C(5)+C(6)*P+C(7)*P(-1)+C(8)*K(-1)
Demand for labor:
W1=C(9)+C(10)*X+C(11)*X(-1)+C(12)*TIME
恒等式,X=CT+I+G； P=X-T-W1；
K=K(-1)+I
endogenous variables:
CT,消费 ; I,投资 ; P:私有部门利润 ;
W1:私有部门工资 ; K,资本存量 ; X:均衡需求 ;
predetermined variables:
T,税收 ; G,政府非工资支出 ; W2,政府工资支出
TIME,时间趋势 ; K(-1); P(-1); X(-1)
Homework:
1,证明阶条件等价于：该方程所不包含的所有变量的数目大于等于模型的总内生变量数减 1。
2,Page236,Ex12.7 and Ex12.8
1iik k g