Josef Perktold
Josef Perktold
What to do if an observed count is zero? related #7011 zero correction in contingency tables meta-analysis, AFAIR, I added several options In those cases above, a fixed number (e.g....
power_proportions_2indep does not support `value` different from equality. Also, not `compare` option, which will become only relevant if non-zero null is supported. see comments at around here https://github.com/statsmodels/statsmodels/issues/8049#issuecomment-1025209120 I partially...
MOVER = Method Of Variance Estimates Recovery An issue search only finds this comment https://github.com/statsmodels/statsmodels/issues/2955#issuecomment-220075363 Zou, G. Y., and A. Donner. 2008. “Construction of Confidence Limits about Effect Measures: A...
(I wanted to add an ENH issue for one sample TOST, but we already have several) ## One-Sample Equivalence tests - `statsmodels.stats.weightstats.DescrStatsW.ttost_mean` - `statsmodels.stats.weightstats.DescrStatsW.ztost_mean` - `statsmodels.stats.weightstats.ztost` combines one and two...
(I don't have an overview and we don't have much) What are the questions and target statistics? currently: meta-analysis based on k 2-sample comparisons stratified contingency table MH test possible...
AFAICS, patsy sets all exterior knots in `BS` at the same points, i.e. lower_bound and upper_bound. https://github.com/pydata/patsy/blob/master/patsy/splines.py#L229 I'm trying to replicate some mgcv functions, and mgcv chooses by default spread...
notebook for recently added functions for 1 and 2 sample poisson rates
Great example collection with "equivalent" code in R and Python (statsmodels, linearmodels, arch) Some examples are not available in packages and coded directly in example, e.g. Tobit https://web.pdx.edu/~crkl/ceR/ This is...
Optimization in discrete models and GLM might not fail with singular or near-singular matrix. We should add a warning text to summary similarly as in linear models OLS, ... example...
quantile residual seems to be a "newer" alias for normal scores, i.e. use ppf to transform to normal distribution related: uniform scores for count distributions, see #1537 #2027 and maybe...