For the latest version of JMP Help, visit JMP.com/help.


Publication date: 06/21/2023

Fit Statistics and Model Summary

In the Generalized Linear Mixed Model report, the Fit Statistics and Model Summary sections contain information that describes the model you have fit.

Fit Statistics

The columns in the Fit Statistics table depend on if there are random effects in the model.

No Random Effects

If there are no random effects, the table contains the following statistics for the model fit:

-2 Log Likelihood

Twice the negative of the natural logarithm of the likelihood function for the model. See Likelihood, AICc, and BIC. Use the log-likelihood for model comparisons in which the fixed, random, and repeated effects differ in any of the models.

Pearson Chi-Square

The Pearson chi-square statistic for the model.

Pearson Chi-Square / DF

The Pearson chi-square statistic for the model divided by the degrees of freedom for the model.

Tip: The Pearson Chi-Square / DF value can be used to evaluate if overdispersion is present in the data versus the model. If this value is much greater than 1, that is evidence of overdispersion. Specifically, if you fit a model with no random effects and the Pearson Chi-Square / DF is much greater than 1, the overdispersion could be due to a missing random effect. See Additional Example of the Generalized Linear Mixed Model Personality.

Random Effects

If there are random effects, the table contains the following statistics for the model fit:

-2 Residual Pseudo-Likelihood

Twice the negative of the natural logarithm of the residual pseudo-likelihood function for the model.

Generalized Chi-Square

The generalized chi-square statistic for the model.

Generalized Chi-Square / DF

The generalized chi-square statistic for the model divided by the degrees of freedom for the model.

Tip: The Generalized Chi-Square / DF value can be used to evaluate if overdispersion is present in the data versus the model. If this value is much greater than 1, that is evidence of overdispersion. See Additional Example of the Generalized Linear Mixed Model Personality.

Model Summary

The Model Summary table contains the following information about the model:

Response

The column assigned to the Y role in the Fit Model launch window.

Distribution

The Distribution selected in the Fit Model launch window.

Probability Model Link

(Appears only when the Distribution is Binomial.) The link function for the model for the probability.

Mean Model Link

(Appears only when the Distribution is Poisson.) The link function for the mean.

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