ENH: Support ARMA(0,0) models. #1370

Merged
merged 2 commits into from Feb 7, 2014

Projects

None yet

2 participants

@jseabold
Member
jseabold commented Feb 7, 2014

See #1262. This has the same changes with the history rewritten a bit.

@jseabold
Member
jseabold commented Feb 7, 2014

I'm going to go ahead and merge this, so it doesn't sit any longer. cc @bashtage @josef-pkt

@jseabold
Member
jseabold commented Feb 7, 2014

Nevermind. AIC and BIC don't match gretl. They use OLS for this case and return an "estimated using least squares ( = MLE)." Should we just return a regression results instance if the order is (0,0)?

@jseabold
Member
jseabold commented Feb 7, 2014

I guess it could be nice to have the prediction dates magic...

@jseabold
Member
jseabold commented Feb 7, 2014

To muddy the waters, Stata returns the same AIC and BIC we use ie., the ARMA definitions.

@jseabold
Member
jseabold commented Feb 7, 2014

R's base arima uses the same AIC as we do (the ARMA one), and I suppose we want these to be comparable, so we should just stick with what we have in this PR I think.

@jseabold jseabold merged commit baff17b into master Feb 7, 2014

1 check passed

default The Travis CI build passed
Details
@jseabold jseabold deleted the arima-order-0-0 branch Feb 7, 2014
@bashtage
Contributor

I think the IC situation should probably be re-considered someday - but in a different PR after discussion.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment