The General Loglinear Model

Loglinear models are "ANOVA-like" models for the log-expected cell counts of crosstabulation tables.

Factors. Factors are the categorical variables that define the crosstabulation table.

Covariates. Scale predictors can be added as covariates in the model. Within cells of the crosstabulation table, the mean covariate values of cases in the cell are used to model the cell counts.

Cell Structure. The cell structure variable allows you assign weights to cells, which can be helpful for imposing a particular structure on the crosstabulation table, fitting rate models such as Poisson regression or certain survival models, or table standardization. See the topic Paired Data for more information.

Contrasts. You can specify a set of contrast variables to test the differences between model effects.

Distributional Assumptions. With a Poisson distribution, the model inference does not depend on sample size and the event a new observation will fall in a given cell is independent of the counts of other cells. With a Multinomial distribution, the inference does depend on sample size and the event a new observation will fall in a given cell is dependent on the counts of other cells.

Next