I would like to use LMTP with a count outcome but it appears that the package is internally scaling Y before estimating the outcome regression. As a result, I cannot use some of the algorithms that are commonly used for count outcomes (like a zero-inflated negative binomial model). Would it be possible to scale Y and Q after estimating the outcome regression?
Hello,
I would like to use LMTP with a count outcome but it appears that the package is internally scaling Y before estimating the outcome regression. As a result, I cannot use some of the algorithms that are commonly used for count outcomes (like a zero-inflated negative binomial model). Would it be possible to scale Y and Q after estimating the outcome regression?
Thank you, Lauren