Open lizzieinvancouver opened 4 years ago
The European dataset with most species is Laube14a and North America is Flynn18. Totals 593 rows of data which may be too small for our models. I ran 6000 iter, 5000 warmup adapt_delta .99 and still had 12 divergent transitions, bad effective sample sizes and Rhats and missing Bayes fractions (whatever that is).
Thoughts on how to proceed?
@dbuona I can take a look; can you point me to the code?
@dbuona Compare answers using just two datasets: one from each continent.... use a study with max species on each continent