DSC 410/510 - Multivariate Statistical Methods

Assignment 16 (deadline: 4pm on Monday, November 29)

Practical matters

Illustration of Correspondence Analysis from p565-570:

• Reproduce the analysis described in the book with the help of the "Illustration of Correspondence Analysis from p565-570" handout. This provides guidance on using SAS to run a correspondence analysis of hatco data (available in the "hatcocorresp.xls" spreadsheet on the data page).
• Select Graph > Axis Variables > Exchange X and Y to swap the axes so that the map looks more like the one on p567.
• Send an e-mail to ipardoe at lcbmail.uoregon.edu with answers to the following questions:
  1. Using the perceptual map, which four firms are Hatco perceived to be most similar to (with regard to the eight attributes measured)?
  2. On the perceptual map there are four attributes close to Hatco. However, only two of these have large positive chi-square deviations suggesting that Hatco scores particularly highly on these attributes (relative to the competing firms). What are these two attributes and what are their chi-square deviations?
  3. On the perceptual map there are two attributes far from Hatco. Check whether these have large negative chi-square deviations to suggest that Hatco scores particularly poorly on these attributes (relative to the competing firms). What are these two attributes and what are their chi-square deviations?

This assignment is worth 15 points rather than the usual 10.

Review questions

We may discuss some of the following questions in class on Tuesday November 30. I will also provide suggested solutions to allow you to check your answers.

• Page 571: questions 1, 5, 7, and 8.