各位牛人,我想问下面这个问题:
我用了msgps包做变量选择、这个包可以用来做adaptive lasso的选择变量、
输入一下命令:
X <- matrix(rnorm(100*8),100,8)
beta0 <- c(3,1.5,0,0,2,0,0,0)
epsilon <- rnorm(100,sd=3)
y <- X %*% beta0 + epsilon
y <- c(y)
fit4 <- msgps(X,y,penalty="alasso",gamma=1,lambda=0)
summary(fit4)
出来结果如下:
Call: msgps(X = X, y = y, penalty = "alasso", gamma = 1, lambda = 0)
Penalty: "alasso"
gamma: 1
lambda: 0
df:
tuning df
[1,] 0.0000 0.0000
[2,] 0.1473 0.1371
[3,] 0.2948 0.2743
[4,] 0.4422 0.4115
[5,] 0.6116 0.5757
[6,] 0.7938 0.7497
[7,] 0.9763 0.9249
[8,] 1.1588 1.1009
[9,] 1.3413 1.2782
[10,] 1.5238 1.4567
[11,] 1.7062 1.6365
[12,] 1.8888 1.8182
[13,] 2.0714 2.0018
[14,] 2.2538 2.1878
[15,] 2.4363 2.3772
[16,] 2.6189 2.5713
[17,] 2.8013 2.7728
[18,] 3.2277 3.1959
[19,] 4.2912 4.1417
[20,] 8.6774 7.8223
tuning.max: 8.695
ms.coef:
Cp AICC GCV BIC
(Intercept) -0.1826 -0.1824 -0.1828 -0.1782
V1 2.9672 2.9668 2.9676 2.9238
V2 2.3078 2.3070 2.3091 2.2192
V3 0.0000 0.0000 0.0000 0.0000
V4 0.0000 0.0000 0.0000 0.0000
V5 2.0431 2.0423 2.0439 1.9690
V6 0.0000 0.0000 0.0000 0.0000
V7 -0.2087 -0.2041 -0.2128 0.0000
V8 0.0000 0.0000 0.0000 0.0000
ms.tuning:
Cp AICC GCV BIC
[1,] 3.562 3.549 3.575 2.877
ms.df:
Cp AICC GCV BIC
[1,] 3.5 3.487 3.511 2.860
我想要的是准则GCV或者BIC的系数,应该如何提取?