Skip to content

g_ar1/p_ar1 very sensitive to initial guess. #131

@lisa-gm

Description

@lisa-gm

When running p_ar1/run.py or g_ar1/run.py with reference theta's: tau = 1 and phi = 0.9, it seems to converge fine for tau & phi > 0.5 but if either is smaller I usually get a matrix not positive definite issue. Check transformations if they cause numeric instability or are generally wrong? To keep in mind that phi = 0.5 will lead to a zero after logit transformation.

Metadata

Metadata

Assignees

No one assigned

    Labels

    Unexpected BehaviorThe issue describe a bug encountered at runtime. The issue should describe how to reproduce the bug.

    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