This preview shows pages 1–8. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: Will Landau November 20, 2011 STAT 231 Homework 11 Solutions 1 To get a normal probability plot of the standardized residuals, I first put the stan dardized residuals in one column of a table and go to Analyze > Distribution: 2 I drag my column name into the Y, Response box and hit OK: 3 I click on the red triangle next to Standardized Residuals in my output and select a Normal Quantile Plot: 4 Now I make the display easier to read. I click and drag the plot window so that the plot appears bigger. Also, I highlight all the data in my table, right click the plot area, and select a bigger marker under the Row Markers submenu: 5 This points deviate severely from a straight line, which show that the standardized residuals are highly nonnormal. 6 31. a. This proportion of variation is our R 2 value of 98.8%. b. The directions for homework 11 say to get the unadjusted R 2 values instead of the adjusted ones. For the cubic model, R 2 = 0 . 980 from the Minitab output in the problem statement. For the quadratic model, the problem statement (directions to part B) gives an R 2 value of 0.780. Informally, this suggests that the cubic model provides a better fit than the quadratic one, although we would need a hypothesis test to determine if the difference in R 2 values is significant.values is significant....
View
Full
Document
This note was uploaded on 02/11/2012 for the course STAT 231 taught by Professor Staff during the Fall '08 term at Iowa State.
 Fall '08
 Staff
 Probability

Click to edit the document details