The accompanying data was taken from the article “Applying Stepwise Multiple Regression Analysis to the Reaction of Formaldehyde with Cotton Cellulose” (Textile Research J., 1984: 157–165). Thedependent variable is durable press rating, a quantitative measure of wrinkle resistance, and the four independent variables are formaldehyde concentration, catalyst ratio, curing temperature, and curing time, respectively.
a. Fitting the model with the four independent variables as predictors resulted in the following Minitab output. Does the fitted model appear to be useful?
b. Estimate, in a way that conveys information about precision and reliability, the average change in durability press rating associated with a 1-degree increase in curing temperature when concentration, catalyst ratio, and curing time all remain fixed.
c. Given that catalyst ratio, curing temperature, and curing time all remain in the model, do you think that formaldehyde concentration provides useful information about durable press rating?
d. Now consider models based not only on these four independent variables but also on secondorder predictors (four xi 2 predictors and six xi xj predictors). Use a statistical computer package to identify a good model based on this candidate pool of predictors.