PR3_2010 - double-click on one of the levels of Target_B,...

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

View Full Document Right Arrow Icon
DSCI 4520/5240 – Data-Based Decision Support Systems PR3 Data: DONOR_RAW Analysis: Imputation, Transformations, and Logistic Regression Due: Oct 13, in class Instructions: 1. Continue the analysis we started in PR1. Make sure you entered Prior Probability information as shown in Getting Started with SAS Enterprise Miner (GSEM) text (p. 39) and Profit Matrix information (p. 40) correctly. Then follow the analysis steps described in the GSEM text, pp. 103-124 to fit a logistic regression model, after imputing and transforming the variables as sown on the text. 2. Turn in a printed word document with (i) a brief one-paragraph description of the project problem, (ii) a brief one-paragraph outline of your analysis including a screen shot of your final Diagram as shown on p.121, and (iii) screen shots of various analysis steps as listed below. Screen Shots: 1. Histograms of the transformed predictor variables, as shown on p. 124, after you
Background image of page 1
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: double-click on one of the levels of Target_B, as instructed on p. 123. Expand the individual windows and presented each variable on a separate graph. After transferring the graphs to Word, make them small enough so that all 20 graphs can fit in two pages. 2. The Logistic Regression results output, showing the odds ratio estimates of the FINAL model (do not exceed one page). Questions: 1. Look at the graphs in screen #1. Some of them show that the percentage for Target_B increases (or decreases) monotonically with the change of the input variable. Some others appear to have a plateau (the percentage for Target_B across the input variable range increases, then decreases again). What is the implication of this observation regarding polynomial orders for the respective inputs? Have we included any polynomial orders? What should we do? 2. Interpret each odds ratio coefficient included in the final model....
View Full Document

This note was uploaded on 04/10/2011 for the course DSCI 4520 taught by Professor Staff during the Spring '08 term at North Texas.

Ask a homework question - tutors are online