Greatings for all,

Suppose we are using the Latent Variable Formulation (i.e. not the Generalized Linear Modeling) for a multiple regression model :

1)Does the Tobit model can be used if the dependent variable is not continuous but is a count variable (e.g. the number of chidren per family) ?

2) Does the Tobit model can be used if the dependent variable is not continuous but is a polichotomous categorical variable (e.g. the eyes color) ?

Thanks in advance.