Skip to main content

New answers tagged

1 vote
Accepted

Relating animal sightings with land cover - poisson, negative binomial, zero inflated and then LOST

A couple things to help you out. @whuber's comment is a good one but assuming this is all the data you have we can assume your analytic task is to "determine whether there is a relationship ...
ischmidt20's user avatar
0 votes

Negative log-likelihood, high BIC, high R-squared, low error, using a difference-in-differences (DiD) methodology

Beyond @PeterFlom's excellent answer about the relative nature of metrics like log-likelihood and BIC, you're missing a critical point: never use a Poisson model without considering the possibility of ...
Ben Bolker's user avatar
  • 44.2k
2 votes

Negative log-likelihood, high BIC, high R-squared, low error, using a difference-in-differences (DiD) methodology

Log likelihood and BIC (and other similar measures) cannot be evaluated on their own. They are useful for comparing models. See, e.g. this thread. If you divide imports and exports by (say) 1000 I ...
Peter Flom's user avatar
  • 124k
2 votes

Poisson regression given multiple predictors on a repeating ID variable

There are roughly 30K ZIP codes in the US, though not all of these have people in them who are at risk of dying. Let's say you have 20K ZIPs with people at risk your data in your data. You seem to ...
dimitriy's user avatar
  • 36.6k
2 votes

Underdispersion handled with negative binomial distribution?

No I don't believe that the standard negative binomial distribution could ever be a valid model for under dispersed data, as I believe that $\theta < 0$ produces an invalid variance function. ...
den173's user avatar
  • 66

Top 50 recent answers are included