Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

add unbiased distribution fitting #914

Open
oxinabox opened this issue Jun 11, 2019 · 3 comments

Comments

Projects
None yet
4 participants
@oxinabox
Copy link

commented Jun 11, 2019

And where defined,
it is probably the better default for fit to redispatch to than fit_mle.
It also might be worth considering fit to take an argument as to if to fit for an unbiased esimator or to fit as a maximum likelyhood estimator.

@matbesancon

This comment has been minimized.

Copy link
Member

commented Jul 10, 2019

sounds ok to replace fit_mle with an unbiased estimation when available, people who want MLE will call it specifically

@matbesancon matbesancon added this to the 1.0 milestone Jul 10, 2019

@simonbyrne

This comment has been minimized.

Copy link
Contributor

commented Jul 10, 2019

You can't really have an "unbiased distribution estimator", all you can really say is that certain functions of the estimated distribution (such as the first and second moments) are unbiased.

@mschauer

This comment has been minimized.

Copy link
Member

commented Jul 10, 2019

That, and being unbiased is only important in particular situations, e.g. in pseudo-marginal Metropolis–Hastings algorithm, otherwise you'd willingly trade bias and variance.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
You can’t perform that action at this time.