R的glmnet包
- glmnet(x=matrix_x1,y=matrix_y1,alpha=1,family="gaussian");
- cv.glmnet(matrix_x1,matrix_y1,type.measure="mse",alpha=1,family="gaussian");
Df %Dev Lambda
[1,] 0 0.00000 0.7038000
[2,] 1 0.08548 0.6413000
[3,] 1 0.15650 0.5843000
[4,] 1 0.21540 0.5324000
[5,] 2 0.26510 0.4851000
9 x 1 sparse Matrix of class "dgCMatrix"
(Intercept) -0.88
ab_Jsw 0.07
ab_ACE -0.42
ab_MPD .
ab_mean .
ab_var 0.00
ab_skew -0.08
ab_kurt .
Type 0.63
只有输出预测变量的系数,没有标准误。如何能输出OLS(如下)的预测变量的标准误,还是本身就不能得到这个数据呢。
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -0.44197 0.12736 -3.470 0.001041
ra_Jsw -0.10 0.12400 -0.868 0.389508
ra_ACE -0.04 0.10319 -0.425 0.672593
ra_MPD -0.05 0.14043 -0.383 0.703143
ra_mean 0.44 0.12582 3.521 0.000894 **
ra_var -0.62 0.30266 -2.062 0.044091 *
ra_skew 0.37 0.43590 0.855 0.396201