[問題] Logistic regression more robust to outliers than LDA
在 The Elements of Statistical Learning一書裡,
作者於p105比較logistic regression(LR)與linear discriminant analysis(LDA)提到:
"...observations far from the decision boundary are down-weighted by
logistic regression..."
但小弟對於這裡的down-weighting一直百思不解.由於LR和LDA有相同
的regression form,所以down-weighting應該不是因為log函數對
大數的compression而來.那麼,到底直觀上或數學上,LR為何會對outliers
比較robust了?
謝謝!
--
※ 發信站: 批踢踢實業坊(ptt.cc)
◆ From: 168.122.81.155