在之前的代碼

lasso1.cv = cv.glmnet (X, Y, alpha = 1, nfolds = 10)

中，alpha = 1為lasso，alpha=0為嶺回歸，介于0和1之間時(shí)表示模型介于嶺回歸和lasso之間，統(tǒng)稱為Elastic Net，對(duì)其進(jìn)行循環(huán)，找到最小的lambda值

final = c()

for (i in 1:100){

? set.seed(1)

? fit4.cv = cv.glmnet (X, Y, alpha = (i/100), nfolds = 10)

? this_fit = c (i/100, fit4.cv$lambda.min, min(fit4.cv$cvm))

? final = rbind (final, this_fit)

}

final [which.min (final[, 3]), ]

由于lasso和嶺回歸對(duì)自變量的單位敏感，在之前先對(duì)數(shù)據(jù)進(jìn)行標(biāo)準(zhǔn)化，運(yùn)用robustHD程序包里standardize()函數(shù)