Dear all,
I'm working on la logistic model with rare event cases. I have used the Larry King approach described in the paper"Logistic Regression in Rare
Events Data", so I did an undersampling and applied a correction, but one of my jurors says "down sampling isn't necessary.But you can use all the observations and then weight to account for low prevalence (bayes posterior)."
I'm a lot confused, I have been looking for bayes posterior adjust in logistic rare events cases but I have found nothing. I really appreciate if any of you can refer me to any paper or book.
Thank you,
Angelica,
_________________________________________________________________
Click less, chat more: Messenger on MSN.ca
http://go.microsoft.com/?linkid=9677404
|