-
-
Notifications
You must be signed in to change notification settings - Fork 411
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
KDE: Display quantiles #479
Conversation
Should we allow for quantiles other than quartiles? If so how the user should pass those options?
|
List of numbers between 0 and 100. Or quantiles => list of numbers between 0 and 1. Also what algorithm should we use for quantiles etc? Data or kde results? |
Good question. I guess using the estimated density is more consistent. |
Also, would there be any potential to define ... "symmetric" quantile, e.g. distribution which is constrained and MAP is on one of the constrain? E.g. 1-tail / 2-tail / negative 1-tail |
Not sure I truly understand what these "symmetric" quantiles are |
True, I probably don't understand them either. I think it similar if we split data based on the density (starting from top, moving horizontal line down until q-quantile of the density is inside the density limits... |
In that case I vote for leaving the "symmetric" quantiles for another PR. |
Good choice |
May I ask for a test? I can write one if you're strapped for time |
Sure! I can add a test, but not until next Tuesday. So if you want to add it sooner than that, please go ahead. |
LGTM |
@ahartikainen are you also OK with this? |
Lgtm |
This add the option to get a KDE segmented into quartiles (or more general quantiles). There are still a few things to fix and improve