Skip to content

Conformal Prediction (For Hierarchical Time Series) #10

@katiavas

Description

@katiavas

My question is can I sum up conformal bounds so that they maintain their validity?

More context i needed: I am using Ensemble batch prediction intervals (EnbPI) to output conformal upper and lower bounds for my time series data. I want to reconcile the upper bound and then the lower bound (input each one in the formula instead of y_pred) with MinTrace(or any other reconciliation method for that matter). Will this affect the validity coverage of my bounds or violate some assumptions? Or it is okay to sum up bounds like this ?

image

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions