Is it possible to use different covariates for the continuous and discrete models?

[daniel-conn17]

I was wondering if it was possible to use different models for the continuous and discrete models in zlm. I would like to fit such a model.

For example, one could have model that looks as follows based off of the documentation for zlm:

figure out of MAST allows you to use different covariates for the discrete and

continuous models

data<- data.frame(x=rnorm(500), z=rbinom(500, 1, .3), r=rnorm(500)) logit.y <- with(data, x2 + z2); mu.y <- with(data, 10+10x+10r + rnorm(500)) y <- (runif(500)<exp(logit.y)/(1+exp(logit.y)))*1 y[y>0] <- mu.y[y>0] data$y <- y

It would be nice if a call such as "fit <- zlm(y ~ x + r | x + z, data)" would work. Right now, it's just fitting a model with an intercept, when I try this.

[amcdavid]

No, there's no support for this currently, nor any plan for this. The package VGAM supports something like this, though I am not sure if they have the zero-inflated log-normal as a family.

[daniel-conn17]

Ok, thank you anyway. I'll check out VGAM.