Josef Perktold
Josef Perktold
In PR #849 I merged Tyler's enhancements to tukeyhsd into master (see also #809 #789 #824 ) some things left to do - another round of proofreading docstrings, and cleaner...
rates.rst wrong function names https://www.statsmodels.org/dev/generated/statsmodels.stats.rates.test_poisson_2indep.html - [x] `alternative` description is only for `ratio` https://www.statsmodels.org/dev/generated/statsmodels.stats.rates.tost_poisson_2indep.html - [x] docstring not updated for "compare" option, i.e. support for "diff" - [x] same for...
What notebooks do we need to add, especially for recently added features? I don't have an overview and most of my private notebooks are dev notebooks with extra experiments or...
We could add excess dispersion `scale` option to cov_type in discrete models similar to GLM. This is mainly relevant for count models, Poisson, NB, GP. I don't remember the details...
I have seen several examples, e.g. on stackoverflow, where users do not set the GLM-negbin parameter. I guess users often assume that the extra parameter is estimated. A warning if...
Just trying to get a rough overview what I'm planning for 0.14 - diagnostics, score tests - missing pieces in discrete (avoiding PRs that get too large or widespread). -...
(non-TSA models) #1353 similar issue, more general extra results methods are available for some models, e.g. `get_influence`, `get_diagnostic`, `get_prediction`, `get_distribution`. `get_margeff` right part of table are tools, `sf, hf` is...
rebased version of #2677 removed last commit which was compat fixes for numpy < 1.9 (in percentile) status not clear see old comments https://github.com/statsmodels/statsmodels/pull/2677#issuecomment-226978367 predict and results info incomplete or...
I wrote unit tests for vectorized tost_proportions_2indep. The unit tests fail with incorrect numbers for odds-ratio with method = 'logit-smoothed' The unit tests for all other `compare` and methods pass....
(I don't have an overview and I didn't find an overview issue) What functions do we have for hypothesis tests and related inference outside of models? What functionality should we...