[Bug]: offset for length and agecomp using Dirichlet was incorrect for obs with males and females #362
Description
Describe the bug
offset_a(f, i) += gammln(nsamp_a(f, i) + 1.) -
sum(gammln(1. + nsamp_a(f, i) * obs_a(f, i)(z1, z2)));
The first part of this eq was being added once for females and then again for males.
This occurs only with Dirichlet and only when the comp sample had both males and females
It was detected as a problem when comparing the logL to SizeComp method which does females and males in one calculation.
To Reproduce
per above
Expected behavior
expected length comp approach and sizefreq approach to be able to give identical result when configured the same
Screenshots
No response
Which OS are you seeing the problem on?
No response
Which version of SS3 are you seeing the problem on?
3.30.19 and earlier
Additional Context
No response