statsmodels / statsmodels/statsmodels
FAQ-D: use_t choosing between normal and t distribution
Nobody has claimed this yet.
- 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
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- 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