Published inData Science CollectiveHow Much Value Survives Over Time? A Value-Based Survival CurveA generalized version of Kaplan-Meier allows to model a continuous value (like money) instead of a binary signal (like survival)2d ago22d ago2
Published inData Science CollectiveHow to Lose Money With “Statistically Significant” DecisionsMany decision-makers rely on statistical significance — but this doesn’t always translate to value. Here’s a more effective framework for…Mar 1217Mar 1217
Published inData Science CollectiveHow to Test if Your Model’s Probabilities Are Good (Enough)Finding a meaningful baseline for the calibration error of any classification modelFeb 186Feb 186
Published inTDS ArchiveThink Correlation Isn’t Causation? Meet Partial CorrelationDespite being so powerful, partial correlation is perhaps the most underrated tool in data scienceJan 819Jan 819
Published inTDS ArchiveWhy “Statistical Significance” Is PointlessHere’s a better framework for data-driven decision-makingDec 1, 202440Dec 1, 202440
Published inTDS ArchiveNon-Linearity: Can Linear Regression Compete With Gradient Boosting?Linear models can handle non-linear relationships thanks to data pre-processing. But how close can they get to more sophisticated models?Oct 11, 202411Oct 11, 202411
Published inTDS ArchiveCausality in ML Models: Introducing Monotonic ConstraintsMonotonic constraints are key to making machine learning models actionable, yet they are still quite unusedSep 6, 202413Sep 6, 202413
Published inTDS ArchiveWhat’s Wrong With R-Squared (And How to Fix It)Even if you think you are using R-Squared out-of-sample, you are not. Here is whyAug 7, 202419Aug 7, 202419
Published inTDS ArchiveForget Statistical Tests: A/B Testing Is All About SimulationsHow simulations outperform traditional stats in that they are easier to understand, more flexible, and economically meaningfulJul 4, 202417Jul 4, 202417
Published inTDS ArchiveHypothesis Testing Explained (How I Wish It Was Explained to Me)Most resources focus on things like Confidence and Power. But they don’t really matter: here is what you should care aboutMay 13, 20249May 13, 20249
