Consider the all-possible-regressions analysis of Hald's cement data in Example 10.1. If the objective is to develop a model to predict new observations, which equation would you recommend and why?

Example 10.1

Transcribed Image Text:

Hald [1952] presents data concerning the heat evolved in calories per gram of cement (y) as a function of the amount of each of four ingredients in the mix: tri- calcium aluminate (x), tricalcium silicate (x2), tetracalcium alumino ferrite (x3), and dicalcium silicate (x4). The data are shown in Appendix Table B.21. These reflect quite serious problems with multicollinearity. The VIFs are: x1:38.496 x2: 254.423 x3: 46.868 x4: 282.513 We will use these data to illustrate the all-possible-regressions approach to vari- able selection.