Josef Perktold
Josef Perktold
a possible idea for power and sample size computation "breadth instead of depth" Instead of getting good power computation for individual tests, we could add a large collection of simplified...
penalizing categorical regressors parking two references Gertheiss, Jan, and Gerhard Tutz. 2009. “Penalized Regression with Ordinal Predictors.” International Statistical Review 77 (3): 345–65. https://doi.org/10.1111/j.1751-5823.2009.00088.x. Tutz, Gerhard, and Jan Gertheiss. 2016....
while looking at #9657 I don't remember or, more likely, never paid attention to that. The loss functions are parameterized for pearson residuals, but the quadratic loss, least squares uses...
an article I saw by chance (random walk with drift in literature search) Zhao, Jun, Yingyu Chen, and Yi Zhang. 2018. “Expectile Regression for Analyzing Heteroscedasticity in High Dimension.” Statistics...
two statsexchange question why threshold parameters in OrderedModel differ from SPSS and R. https://stats.stackexchange.com/questions/592680/why-do-statsmodels-python-and-mass-r-output-different-intercept-values-for-m we already have helper methods for converting threshold params We could also make standard errors available...