statsmodels / statsmodels/statsmodels

FAQ-D: use_t choosing between normal and t distribution

Open
#1,833 2 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

comp-base FAQ
Dominant language
Python
Stars
11.6k
Forks
3.6k
Avg merge
7h 37m
Merged PRs (30d)
96

Description

see also #1759 for user FAQ

How do we implement the choice between normal and t distribution, or between F and chisquare distribution for parameter inference?

Result instances define an attribute use_t that all methods are supposed to use when choosing the distributions.

distribution of parameters is used in:

  • pvalues
  • conf_int
  • t_test
  • wald_test

hardcoded f distribution

  • f_pvalue
  • f_test
  • anova_lm
  • compare_f_test

hardcoded chisquare distribution

  • LR-test like pvalue for llnull, compare_lr_test

not yet checked: Margins, prediction, ...

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start by tracing the Result instances' use_t attribute and the listed inference entry points: pvalues, conf_int, t_test, wald_test, f_pvalue, f_test, anova_lm, compare_f_test, and compare_lr_test. Check the unreviewed Margins and prediction areas as well. Done means the relevant methods consistently choose normal or t, F or chi-square distributions according to use_t.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
data
Issue type
Feature
Difficulty
5/5
Estimated time
Over a week
Activity status
Stale
Clarity
Mostly clear
Newbie friendliness
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.