Data Awesome https://data-awesome.com/blog Writing on data analytics, Alteryx, AI and the career around them, from ten years leading analytics in airline operations. en-us racine.isacar@gmail.com (Isacar Racine) racine.isacar@gmail.com (Isacar Racine) Sat, 01 Aug 2026 00:00:00 GMT https://data-awesome.com/blog/division-prueba-unica-comparar-modelos Una sola división de prueba no basta para comparar modelos https://data-awesome.com/blog/division-prueba-unica-comparar-modelos Una sola división train/test puede hacer que un modelo se vea mejor de lo que es y otro peor, en el mismo experimento. Así se ve eso con números reales. Sat, 01 Aug 2026 00:00:00 GMT racine.isacar@gmail.com (Isacar Racine) model-diagnosticsstatisticsmachine-learning https://data-awesome.com/blog/single-test-split-model-comparison A Single Test Split Isn't Enough to Compare Models https://data-awesome.com/blog/single-test-split-model-comparison One train/test split can make one model look better than it is and another look worse, in the same experiment. Here's what that looks like with real numbers. Sat, 01 Aug 2026 00:00:00 GMT racine.isacar@gmail.com (Isacar Racine) model-diagnosticsstatisticsmachine-learning https://data-awesome.com/blog/regression-trap-index 15 Ways to Misread Your Own Regression Model https://data-awesome.com/blog/regression-trap-index Fifteen ways regression output gets misread, with the symptom, what it actually means, and how to check. Every one came from a real model. Mon, 27 Apr 2026 00:00:00 GMT racine.isacar@gmail.com (Isacar Racine) regressionstatisticsmodel-diagnostics https://data-awesome.com/blog/regression-coefficient-sign-flip The Coefficient That Flipped Sign https://data-awesome.com/blog/regression-coefficient-sign-flip Alone it reads -0.0054. With seven other predictors, +0.0154. Both significant. Why hunting for the guilty variable is the wrong instinct. Mon, 20 Apr 2026 00:00:00 GMT racine.isacar@gmail.com (Isacar Racine) regressionstatisticsmodel-diagnostics https://data-awesome.com/blog/regression-coefficient-not-causal Regression Coefficients Are Not Causal https://data-awesome.com/blog/regression-coefficient-not-causal The smoker coefficient is $24,330. That does not mean quitting saves anyone $24,330, and no amount of significance will make it mean that. Mon, 13 Apr 2026 00:00:00 GMT racine.isacar@gmail.com (Isacar Racine) regressionstatisticsmodel-diagnostics https://data-awesome.com/blog/regression-diagnostics-outliers-multicollinearity The R-Squared Trap: When Cleaning Backfires https://data-awesome.com/blog/regression-diagnostics-outliers-multicollinearity I removed 53 outliers. R-squared rose from 0.818 to 0.856 and the model got worse. Every number that improved ignored the rows I deleted. Mon, 23 Mar 2026 00:00:00 GMT racine.isacar@gmail.com (Isacar Racine) regressionmodel-diagnosticsstatisticsr https://data-awesome.com/blog/simple-linear-regression-assumptions Why Care About Regression Assumptions? https://data-awesome.com/blog/simple-linear-regression-assumptions One broken assumption ruins your coefficients; the other three only ruin your p-values. Knowing which decides whether you scrap the model. Mon, 09 Mar 2026 00:00:00 GMT racine.isacar@gmail.com (Isacar Racine) regressionstatisticsmodel-diagnostics