Special models for categorical variables
Description
`sqtab' contains a set of functions for estimating
loglinear models for square tables such as quasi-independence,
symmetry, uniform association. `mclgen' restructures a
dataframe to enable the estimation of a multinomial logistic
model using the conditional logit program `clogit'. This allows
greater flexibility in imposing constraints on the response
variable. One application is to specify aforementioned models
for square tables as multinomial logistic models with
covariates at the respondent level. `ctab' simplifies the
production of (multiway( percentage tables.