EViews非线性模型参数估计方法步骤
1.新建EViews工作区，并将时间序列X、P1和P0导入到工作区；
2.设定参数的初始值全部为1，其方法是在工作区中其输入下列命令
并按回车键
param c(1) 1 c(2) 1 c(3) 1 c(4) 1
3.估计非线性模型参数，其方法是在工作区中其输入下列命令并按
回车键
nls q=exp(c(1))*x^c(2)*p1^c(3)*p0^c(4)
4.得到结果见table01（91页表3.
5.4结果）（案例一结束）
Dependent Variable: Q
Method: Least Squares
Date: 03/29/15 Time: 21:44
Sample: 1985 2006
Included observations: 22
Convergence achieved after 9 iterations
Q=EXP(C(1))*X^C(2)*P1^C(3)*P0^C(4)
Coefficient Std. Error t-Statistic Prob.
C(1) 5.567708 0.083537 66.64931 0.0000
C(2) 0.555715 0.029067 19.11874 0.0000
C(3) -0.190154 0.143823 -1.322146 0.2027
C(4) -0.394861 0.159291 -2.478866 0.0233
R-squared 0.983631 Mean dependent var 1830.000
Adjusted R-squared 0.980903 S.D. dependent var 365.1392
S.E. of regression 50.45954 Akaike info criterion 10.84319
Sum squared resid 45830.98 Schwarz criterion 11.04156
Log likelihood -115.2751 Hannan-Quinn criter. 10.88992
Durbin-Watson stat 0.672163
（92页表3.5.5结果）（案例二过程）
5.新建EViews工作区，并将时间序列X、P1和P0导入到工作区；
6.设定参数的初始值全部为1，其方法是在工作区中其输入下列命令
并按回车键
param c(1) 1 c(2) 1 c(3) 1
7.估计非线性模型参数，其方法是在工作区中其输入下列命令并按
回车键
nls q=exp(c(1))*(x/p0)^c(2)*(p1/p0)^c(3)
8.得到结果见table02（92页表3.5.5结果）（案例二结束）
Dependent Variable: Q
Method: Least Squares
Date: 03/29/15 Time: 22:14
Sample: 1985 2006
Included observations: 22
Convergence achieved after 4 iterations
Q=EXP(C(1))*(X/P0)^C(2)*(P1/P0)^C(3)
Coefficient Std. Error t-Statistic Prob.
C(1) 5.525965 0.072685 76.02666 0.0000
C(2) 0.533824 0.019785 26.98163 0.0000
C(3) -0.242862 0.134014 -1.812219 0.0858
R-squared 0.982669 Mean dependent var 1830.000
Adjusted R-squared 0.980845 S.D. dependent var 365.1392
S.E. of regression 50.53638 Akaike info criterion 10.80939
Sum squared resid 48524.59 Schwarz criterion 10.95817
Log likelihood -115.9033 Hannan-Quinn criter. 10.84444
Durbin-Watson stat 0.656740
