Prior Probability Constant Adjustment for Probit, Complementary Log-Log?
- From: "Predictor" <predictr@xxxxxxxxxxxxxxxx>
- Date: 9 Aug 2006 10:07:03 -0700
According to this source:
http://www.uic.edu/classes/idsc/ids570/logistic_regr.htm
....the constant of a logistic regression can be adjusted for prior
probabilities different than those in the modeled data:
"Logistic regression implicitly uses prior probability estimates
obtained from the sample. Thus the estimate of B0 will contain a term
ln(p/q), where p and q are the proportions of cases from
Group 0 and Group 1 in the sample. If the appropriate prior
probabilities are instead p' and q',then b0 should be adjusted:
new bo = bo - ln(p/q) + ln(p'/q') ."
What adjustments are used with probit regression and complementary
log-log ("extreme values") regression?
-Will Dwinnell
http://will.dwinnell.com
.
- Follow-Ups:
- Re: Prior Probability Constant Adjustment for Probit, Complementary Log-Log?
- From: David Winsemius
- Re: Prior Probability Constant Adjustment for Probit, Complementary Log-Log?
- Prev by Date: Question relating to order statistics of normal random variables
- Next by Date: Re: Unsolved problems
- Previous by thread: Question relating to order statistics of normal random variables
- Next by thread: Re: Prior Probability Constant Adjustment for Probit, Complementary Log-Log?
- Index(es):
Relevant Pages
|
Loading
