Chapter Contents |
Previous |
Next |

Multiple Regression |

Choose Tables:Collinearity Diagnostics. |

This displays the table shown in Figure 14.7.

**Figure 14.7:** Collinearity Diagnostics Table

When an explanatory variable is nearly a linear
combination of other explanatory variables in the model,
the estimates of the coefficients in the regression
model are unstable and have high standard errors.
This problem is called *collinearity*.
The **Collinearity Diagnostics** table is calculated
using the eigenstructure of the *X*'*X*
matrix.
See Chapter 13, "Fitting Curves," for a complete explanation.
A collinearity problem exists when a component
associated with a high condition index contributes
strongly to the variance of two or more variables.
The highest condition number in this table is **17.0416**.
Belsley, Kuh, and Welsch (1980) propose that a condition
index of 30 to 100 indicates moderate to strong collinearity.

Chapter Contents |
Previous |
Next |
Top |

Copyright © 1999 by SAS Institute Inc., Cary, NC, USA. All rights reserved.