-
Notifications
You must be signed in to change notification settings - Fork 25
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
Underestimated variance in the predictive message #201
Comments
@albertpod |
Just dropping my thoughts on this issue. When you specify We can create a method for computing statistics of such joints. Alternatively, we can provide special treatment for |
I see. To properly fix this issue we probably should reimplement the way |
Following our discussion with @leostenzel and @bartvanerp, I've come across the following issue that occurs with nonmissing (observed) data points when the structured constraint is specified.
This results in
which makes sense given that we can't compute the
mean_cov(q_out_μ)
.To circumvent the error, we should use a different constraint:
which is a correct constraint but will result in underestimated variance in the prediction message (only
γ
will contribute).EDIT: I wouldn't say it's necessarily a bug but it's something we need to think of.
The text was updated successfully, but these errors were encountered: