用jplv7工具箱做SDM模型分析时,为什么直接效应和间接效应有两个(运行结果的最后面),另外直接效应和间接效应是怎么看的?请大神给予指点。
Spatial autoregressive Model Estimates
Dependent Variable = Crime
R-squared = 0.6662
Rbar-squared = 0.6358
sigma^2 = 0.0092
Nobs, Nvars = 49, 5
log-likelihood = 44.291355
# of iterations = 1
min and max rho = -1.0000, 1.0000
total time in secs = 0.0030
No lndet approximation used
***************************************************************
Variable Coefficient Asymptot t-stat z-probability
constant 0.430006 3.385800 0.000710
income -0.918346 -2.779230 0.005449
house value -0.293979 -3.298984 0.000970
W*income -0.533457 -0.943702 0.345322
W*house value 0.250508 1.400432 0.161384
rho 0.422210 2.690226 0.007140
direct indirect total
ans =
-1.0286 -1.4812 -2.5098
-0.2794 0.2028 -0.0766
direct t-stat indirect t-stat total t-stat
ans =
-1.0234 -3.1981 -1.5705 -1.9039 -2.5939 -2.9558
-0.2768 -3.1123 0.2235 0.7631 -0.0533 -0.1624


雷达卡




京公网安备 11010802022788号







