p***l 发帖数: 1775 | 1 multicolinearity解决不了是不是只好转向ridge regresion了?
coef出来之后不知道要做哪些diagonastic |
a****m 发帖数: 693 | 2 依据不同的shrinkage parameter, 检验最后的squared loss before and after
shrinkage of parameter of beta. |
p***l 发帖数: 1775 | 3 多谢楼上回帖,很多书上涉及ridge regression的时候,讲的很粗略。
还有我第一个问题,用stepwise或者其他手段做变量筛选后,检测剩余变量的vif,发
现依然大于10。(center他们也没有用)这时候转向ridge,make sense吗?还有其他
补救措施吗? |
a****m 发帖数: 693 | 4 stepwise can not detect the model complexity, here complexity means that the
amplitude or abundance of collinearity between variable. you may try some
information criteria, such as AIC or Bayesian IC. |