search

lmoffatt / macro_dr

GNU General Public License v3.0
1 stars 1 forks source link

avoid nan logLikelihood in macror #176

Open lmoffatt opened 2 months ago

lmoffatt commented 2 months ago

There should be no reason for having a nan logLikelihood.

Usually what happens is that there is an approximation that fails and blow up to infinity. In this case we can use a more general approximation.

lmoffatt commented 2 months ago

One possibility is to change the logic of macror, directly forbidding nan values, making the algorithm stop with one.

at least we would try this possibility

lmoffatt commented 2 months ago

The nan was caused by a bug in experiment idealization where there was a negative interval.