logisticRegression3_4perPage

logisticRegression3_4perPage - Logistic Regression Data...

Info iconThis preview shows pages 1–3. Sign up to view the full content.

View Full Document Right Arrow Icon

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

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

Unformatted text preview: Logistic Regression Data Mining Prof. Dawn Woodard School of ORIE Cornell University 1 Outline 1 Announcements 2 Logistic Regression 2 Announcements Questions? Remember: the second (other) prelim exam is 12/02 (Wednesday, the second-to-last day of class) 4 Veterans Data Several columns of the veteran’s data: 6 Veterans Data Last time we fit a logistic regression model for TARGET.B, using training data To predict for a test data set, we calculated Pr ( Y = 1 | X 1 , . . . , X P ) If this probability was > . 5, we predicted Y = 1, and otherwise Y = We obtained a matrix of predicted vs. actual outcomes for a test data set: Predict Y = Predict Y = 1 Actually Y = 430 315 Actually Y = 1 327 416 7 Veterans Data Just like for naive Bayes, we can alter the classification threshold With threshold = 0.4: Predict Y = Predict Y = 1 Actually Y = 176 569 Actually Y = 1 118 625 With threshold = 0.6: Predict Y = Predict Y = 1 Actually Y = 629 116 Actually Y = 1 558 185 8 Veterans Data ROC curve: False Positive Rate True Positive Rate 0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0 9 Veterans Data Is our classifier better than the coin-flipper?...
View Full Document

This note was uploaded on 12/23/2009 for the course ORIE 4740 at Cornell.

Page1 / 7

logisticRegression3_4perPage - Logistic Regression Data...

This preview shows document pages 1 - 3. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online