Skip to content

Use distribution distance as a likelihood statistic (Anderson-Darling, Wasserstein, Energy)? #459

@Gabriel-p

Description

@Gabriel-p

See about using the A-D test as a likelihood. Questions:

  1. Is this a valid likelihood?
  2. Does it also fit for the number of points?

Also scipy's distances wasserstein_distance, energy_distance can be used

Related:

How to find out if two datasets are close to each other?

Metadata

Metadata

Assignees

Type

No type
No fields configured for issues without a type.

Projects

No projects

Milestone

No milestone

Relationships

None yet

Development

No branches or pull requests

Issue actions