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

Feature request MvNormal for covellipse #489

Open
ga72kud opened this issue Feb 10, 2022 · 3 comments
Open

Feature request MvNormal for covellipse #489

ga72kud opened this issue Feb 10, 2022 · 3 comments

Comments

@ga72kud
Copy link

ga72kud commented Feb 10, 2022

It would be nice if the second line would also work

N=MvNormal(μ, Σ)

covellipse(N, n_std=2, aspect_ratio=1, label="cov1")

covellipse(μ, Σ, n_std=2, aspect_ratio=1, label="cov1")
@sethaxen
Copy link
Member

I agree, this would be nice.

In general I think this code can be improved.
e.g. We can support trivariate normals by plotting an ellipsoid.
We can also support n-variate normals by allowing the user to specify which slice of the covariance matrix should be plotted.
A corellipse function would be useful as well.
And we could even go so far as to support something like corrplot that plots ellipses for all bivariate slices of the covariance matrix.

Then, for Bayesian modeling, it's useful to view distributions of covariance/correlation matrices. This paper introduces some plots for this: http://www.stat.columbia.edu/~gelman/research/unpublished/Visualization.pdf

@ga72kud
Copy link
Author

ga72kud commented Feb 10, 2022

Thank you! That sounds great!

@ga72kud ga72kud closed this as completed Feb 14, 2022
@sethaxen sethaxen reopened this Feb 14, 2022
@sethaxen
Copy link
Member

@ga72kud I'm re-opening so we can track this feature.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

No branches or pull requests

2 participants