search

drisso / zinbwave

Clone of the Bioconductor repository for the zinbwave package, see https://bioconductor.org/packages/zinbwave
43 stars 10 forks source link

Verbose mode prints wrong value of log likelihood #30

Closed drisso closed 6 years ago

drisso commented 6 years ago

See example here: https://github.com/statOmics/zinbwaveZinger/issues/2

drisso commented 6 years ago

I know what's going on.

In an earlier version of zinbFit we used to update the object m at each iteration. We now only do that at the end of the function. Hence, when at each iteration we compute loglik(m, Y) it is always the initial value.