Dominance Analysis Command for Multilevel Zero-Inflated Negative Binomial Regression

[David Okunlola]

Dear everyone,

Could anyone assist with the Stata command for performing a dominance analysis on a Multilevel Zero-inflated Negative Binomial Regression on Stata 14?

I also welcome any form of guidance from you.

Thank you very much.

[Joseph Luchman]

Hi David,

domin (on SSC; domme, also on SSC, is likely a better fit here but hasn't been tested in V14) could work to conduct dominance analysis but you'd have to find an implementation of the desired multilevel zero-inflated negative Binomial - which I'm not sure would even be straightforward (or possible) to estimate with existing commands in Stata V 16.1 (the version I have access to; e.g., using gsem).

Consider using just the zinb (losing the multilevel element) as that is a model implementable in Stata V14. Would have to write a custom wrapper command to accommodate zinb's inflate() option as well. The structure of such a wrapper command is discussed in domin's help file and may be a tricky programming puzzle to implement should you decide to do so.

- joe

[David Okunlola]

Thank you for your response, Joe