206 followers
RT @predict_addict: @rasbt But if baseline is XGBoost? As @GaelVaroquaux states one should abandon linear models if sufficient data, also L…
RT @predict_addict: @rasbt But if baseline is XGBoost? As @GaelVaroquaux states one should abandon linear models if sufficient data, also L…
@rasbt But if baseline is XGBoost? As @GaelVaroquaux states one should abandon linear models if sufficient data, also Logistic Regression is not automatically a good benchmark because it sometimes produces miscalibrated predictions https://t.co/wxUfao5M6v
@mph_for_doctors MLEはconsistentですが、unbiasedではないですよ。 https://t.co/PhH9bKCvzQ
Bias in odds ratios by logistic regression modelling and sample size http://t.co/6g7UM68EyP