Skip to content

[REFACTOR] Streamline WeightedRegression API #1100

Description

@henrydingliu

Description

WeightedRegression was originally created to help with ldf calculation and mack statistics calculation. We are finding more use for this helper class, e.g. Friedland Chapter 11, and should formally introduce it in the public API. however, the class could use some streamlining prior to adding to the api doc.

Is your feature request aligned with the scope of the package?

  • Yes, absolutely!
  • No, but it's still worth discussing.
  • N/A (this request is not a codebase enhancement).

Describe the solution you'd like, or your current workaround.

here is a list of proposed enhancements, each could probably be its own sub-issue

  • eliminate the xp argument; instead, extract xp directly from the arrays being passed in
  • move all sigma and std_err fills into fit, make fill/interpolation functions private, add argument to class declaration to control fill/interpolation behavior
  • harmonize OLS and OLS_thru_origin
  • improve docstrings and add to api toc

Do you have any additional supporting notes?

No response

Would you be willing to contribute this ticket?

  • Yes, absolutely!
  • Yes, but I would like some help.
  • No.

Metadata

Metadata

Assignees

No one assigned

    Labels

    Type

    No type

    Fields

    Priority

    None yet

    Scope

    None yet

    Effort

    None yet

    Projects

    Status
    Backlog

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions