-
-
Notifications
You must be signed in to change notification settings - Fork 2k
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
Rice distribution is misdefined #3286
Comments
Good catch! |
nbud
added a commit
to nbud/pymc3
that referenced
this issue
Dec 4, 2018
nbud
added a commit
to nbud/pymc3
that referenced
this issue
Dec 5, 2018
twiecki
pushed a commit
that referenced
this issue
Dec 6, 2018
* Fix Rice distribution and add new parametrization (#3286) * fix math error in the docstring introduced by prev commit * code format * Update RELEASE-NOTES.md * Add i1e and i0e.grad * elemwise i0e and i1e, does not work * elemwise i0e and i1e * Rice now accepts tensor parameters * update RELEASE-NOTES.md
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Description of your problem
The Rice distribution
Rice(nu, sigma)
is the distribution ofsqrt(X^2+Y^2)
whereX~N(nu, sigma^2)
andY~N(nu, sigma^2)
, with X and Y independent. This parametrisation is the one notably used in the Wikipedia article on Rice distribution.scipy.stats.rice used a different distribution: shape parameter
b=nu/sigma
and scalesigma
.The current Rice distribution in pymc inconsistently mixes the two parametrisations:
nu
is sometimes the location parameter of the normal distributions, sometimes the shape parameter.I'm writing a PR to fix this.
The text was updated successfully, but these errors were encountered: