Logistic Regression, Prediction and ROC

Ucedubbcswebdavpid 9566224 dt content r id

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: .l0 umr(rdtgm) Note that there might be a warning message “Warning: glm .fit: fitted probabilities num erically 0 or 1 occurred”. This happens because of a problem called quasi-complete separation. You can learn more here or here. The offending variable is X9. We will choose to ignore the warning now. You can try fitting your model without X9. The usual stepwise variable selection still works for logistic regression. caution: this w ill take a very long time. cei.l.tp< se(rdtgm) rdtgmse - tpcei.l0 Or you can try model selection with BIC: cei.l.tp< se(rdtgm,k= rdtgmse - tpcei.l0 lgno(rdttan) o(rwcei.ri)) https://blackboar d.uc.edu/bbcswebdav/pid- 9566224- dt- content- r id- 55868231_2/cour ses/14SS_BANA7046002/notes%284%29.html 2/15 2/17/2014 Log istic Reg r ession, Pr ediction and ROC Prediction and Cross Validation Using Logistic Regression Now suppose there are 2 models we want to test, one with all X variables(credit.glm0), and one with X3, X8 and X11_2(credit.glm1). cei.l1< gmY~X +X +X12 fml =bnma, rdtgm - l( 3 8 1...
View Full Document

Ask a homework question - tutors are online