Skip to content

Normal CDF implementation in stan #4

@pboesu

Description

@pboesu

Currently the likelihood calculations use the stan Phi() function for CDF terms but normal_lpdf() for PDF terms.
Investigate whether there are computational (speed, stability) or consistency gains to be made by using Phi_approx() or normal_(l)cdf() instead.

Metadata

Metadata

Assignees

No one assigned

    Labels

    consistencythis is something that isn't handled consistently across the project

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions