Statistics 512: Problem Set No. 7 Due October 30, 2009 1. For this problem use the CS dataset examined in previous problem sets, and use the model which uses only HSM and HSE as explanatory variables to predict the response GPA. On Homework 6, Problem 3 you examined some visual diagnostics (for your chosen model), which should have include plots of Y vs. X, residuals, etc. Now, additionally examine other diagnostics such as (but not necessarily exclusively) studentized and studentized-deleted residuals, Cook’s D, tolerance or vif, and partial residual plots. Explain any problems such as outliers, highly inﬂuential observations or multicollinearity that these diagnostics point out. (Do not include in your output any tables of values for all 224 individuals. Use plots and verbal summaries instead. You may include values for a few selected individuals if you wish.) Use the dataset from Problem 16.11 described on page 725 of KNNL ( ch16pr11a.dat ). This analysis will be continued on the next homework set, so be sure to save your SAS
This is the end of the preview.
access the rest of the document.