search

bbolker / bbmle

maximum likelihood estimation package
GNU General Public License v3.0
25 stars 13 forks source link

Fixups for #12 #14

Closed andrechalom closed 7 years ago

andrechalom commented 7 years ago

Turns out I hadn't sufficiently tested the changes... sorry for the mess! When running AICc on a list of objects without a "nobs" attribute, the "sapply" on line 114 was creating a list with several NULL entries, instead of a single NULL value. With that, the second null test (line 117) was always false!

I've also changed some sapplys to lapplys to preserve the "df" attributes in the logLik objects, as the original objects passed as "object, ..." may not have df attributes as well.

bbolker commented 7 years ago

thanks. I guess I should have checked more carefully ...