Re: comparisons in text corpora: keywords / CHI square
Marie E. Helt (meh2@dana.ucc.nau.edu)
Fri, 30 Aug 1996 08:44:55 -0700 (MST)
The discussion about chi squared and log likelihood has been very
enlightening. I wondered if anyone could shed some light on the
difference (if there is one) between log likelihood and maximum
likelihood. I have been using VARBRUL 2S, which computes both chi
squared and a maximum likelihood value - however, I have not been able to
find a description of what the latter value represents in any stats
books.
I look forward to hearing from you all.