#岭回归
#linearRidge
我在做分析时,发现数据有严重的多重共线性,所以采用ridge包的linearRidge函数进行拟合,在summary之后虽然得到了各个变量的系数,但是该如何去评价模型的拟合优度呢?(最好能得到R方)
`代码如下:Call:
linearRidge(formula = y_sell ~ y_comment + y_service + y_describe +
y$防晒 + y$是否进口 + y_quality + y_baozhiqi)
Coefficients:
Estimate Scaled estimate Std. Error (scaled) t value (scaled) Pr(>|t|)
(Intercept) 1.450e+03 NA NA NA NA
y_comment 6.241e-01 3.057e+04 9.980e+02 30.626 < 2e-16 ***
y_service 4.128e+04 9.570e+02 6.543e+02 1.463 0.14358
y_describe -4.533e+02 -1.110e+03 5.670e+02 1.958 0.05029 .
y$防晒是 2.889e+02 2.626e+03 9.987e+02 2.630 0.00855 **
y$是否进口是 -2.670e+02 -4.192e+03 1.010e+03 4.152 3.3e-05 ***
y_quality -1.887e+04 -5.246e+02 5.707e+02 0.919 0.35803
y_baozhiqi3年 -1.014e+02 -7.182e+02 9.303e+02 0.772 0.44014
y_baozhiqi3年以上 -1.027e+02 -8.953e+02 9.296e+02 0.963 0.33548
Signif. codes: 0 ‘’ 0.001 ‘’ 0.01 ‘’ 0.05 ‘.’ 0.1 ‘ ’ 1
Ridge parameter: 0.4715623, chosen automatically, computed using 2 PCs
Degrees of freedom: model 4.061 , variance 2.789 , residual 5.334
`