Logistic Regression, Prediction and ROC

# Cost1 is a symmetric cost function ie getting 1 wrong

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

This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: istic Reg r ession, Pr ediction and ROC # # Mdl Ae oe ra pvlebnr.ra .au iomae # 1Mdl 1088 741-0 # oe .29 .8e1 N A # 2Mdl 2077 107-6 # oe .58 .7e0 N A Another library ROCR can also generate ROC curves for you. isalpcae(RC&quot; ntl.akgs&quot;OR) lbayRC) irr(OR pe &lt; peito(rbgm.usml,cei.etY rd - rdcinpo.l1otape rdtts\$) pr &lt; promnepe,&quot;p&quot; &quot;p&quot; ef - efrac(rd tr, fr) po(ef clrz =TU) ltpr, ooie RE https://blackboar d.uc.edu/bbcswebdav/pid- 9566224- dt- content- r id- 55868231_2/cour ses/14SS_BANA7046002/notes%284%29.html 9/15 2/17/2014 Log istic Reg r ession, Pr ediction and ROC Cross v alidation and cost function Refer to lecture slides and Elements of Statistical Learning book (section 7.10) for more advice on cross validation. You should read the cv.glm help to understand how it works. First we define the cut-off probability and the define the cost function. cost1 is a symmetric cost function, i.e. getting 1 wrong cost the same as getting 0 wrong. cost2 is asymmetric cost function (we are...
View Full Document

## This document was uploaded on 03/18/2014.

Ask a homework question - tutors are online