when we run binary outcome variables in a Growth Curve Analysis with no exogenous variables, some of the bivariate frequency tables have zero cells in the bivariate table. This creates the program to stop. Is there any option to circumvent this problem like using "ridge constant"?
Please send your input, data, output, and license number to firstname.lastname@example.org. Categorical data methods are not developed for use with bivariate tables with zero cells so you may need to collapse your data.