posterior SDs are all zero for Poisson likelihood

stephens999 / ashr

An R package for adaptive shrinkage

GNU General Public License v3.0

80 stars 36 forks source link

posterior SDs are all zero for Poisson likelihood #98

Open willwerscheid opened 5 years ago

willwerscheid commented 5 years ago

I think fixing this is as simple as changing alpha = outer(-x, m$a,FUN="+")/s to alpha = outer(x, -m$b,FUN="+")/s in comp_postmean2 (since we send -b and -a to e2truncgamma) but I need to check that this change wouldn't break other likelihoods.