Publication date: 04/12/2021

This section contains statistical details for the Whole Model Test report.

The Chi-Square statistic is sometimes denoted G2 and is defined as follows:

where the summations are over all observations instead of all cells.

The ratio of this test statistic to the background log-likelihood is subtracted from 1 to calculate R2. More simply, RSquare (U) is computed as follows:

using quantities from the Whole Model Test report.

Note: RSquare (U) is also known as McFaddenâ€™s pseudo R-square.

Want more information? Have questions? Get answers in the JMP User Community (community.jmp.com).