Publication date: 07/08/2024

Whole Model Test

In the Generalized Linear Model Fit report, the Whole Model Test table shows tests that compare the whole-model fit to the model that omits all the regression parameters except the intercept parameter. It also contains two goodness-of-fit statistics and the AICc value to assess model adequacy.

The Whole Model Test table shows these quantities:

Model

The model labels.

Difference

The difference between the Full model and the Reduced model. This model is used to measure the significance of the regressors as a whole to the fit.

Full

The complete model that includes the intercepts and all effects.

Reduced

The model that includes only the intercept parameters.

–LogLikelihood

The negative log-likelihood for the respective models. See Likelihood, AICc, and BIC.

Note: When the Overdispersion Tests and Intervals option is selected in the launch window, the -LogLikelihood value is calculated using the quasi-likelihood approach.

L-R ChiSquare

The likelihood ratio chi-square test statistic for the hypothesis that all regression parameters are zero. The test statistic is computed by taking twice the difference in negative log-likelihoods between the fitted model and the reduced model that has only an intercept.

DF

The degrees of freedom (DF) for the Difference between the Full and Reduced model.

Prob>ChiSq

The probability of obtaining a greater chi-square value if the specified model fits no better than the model that includes only an intercept.

Goodness of Fit Statistic

The two goodness-of-fit statistics: Pearson and Deviance.

ChiSquare

The chi-square test statistic for the respective goodness-of-fit statistics.

DF

The degrees of freedom for the respective goodness-of-fit statistics.

Prob>ChiSq

The p-value for the respective goodness-of-fit statistics.

Overdispersion

(Appears only when the Overdispersion Tests and Intervals option is selected in the launch window.) An estimate of the overdispersion parameter. See Statistical Details for the Generalized Linear Model Personality.

AICc

The corrected Akaike Information Criterion. See Likelihood, AICc, and BIC.

Note: When the Overdispersion Tests and Intervals option is selected in the launch window, the AICc calculation does not include the overdispersion parameter.

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