PinnedPublished inData Science CollectiveTeach Your GBM to Extrapolate with Model StackingBackgroundApr 4A response icon6Apr 4A response icon6
Published inPython in Plain EnglishDo Opposites Attract? A Data Science Love StoryIs combining gradient-boosted trees with linear models a silver bullet for extrapolation or a memory-hungry nightmare?Mar 11A response icon1Mar 11A response icon1
Published inPython in Plain EnglishA Cautionary Tale: Why You Might Want to Calibrate Your GBM ModelsA real-world example of some of the dangers lurking below the surface of a “good” GBM model.Feb 10A response icon3Feb 10A response icon3
Published inTDS ArchiveHow Reliable Are Your Time Series Forecasts, Really?How cross-validation, visualisation, and statistical hypothesis testing combine to reveal the optimal forecasting horizonApr 5, 2024A response icon7Apr 5, 2024A response icon7
Published inPython in Plain EnglishTime series residuals: gold in them thar hillsHow model crumbs provide more insight than you might thinkMar 15, 2024A response icon1Mar 15, 2024A response icon1
Published inTDS ArchiveFalse Prophet: Lightning Strikes TwiceInsights from incorporating external weather data into a time series regression inspired by Meta’s ProphetMar 1, 2024A response icon1Mar 1, 2024A response icon1
Published inTDS ArchiveHybrid Models for Time Series RegressionUsing multiple model forms to capture and forecast the components of complex time seriesJan 13, 2024A response icon1Jan 13, 2024A response icon1
Published inTDS ArchiveFalse Prophet: Comparing a Regression Model to Meta’s ProphetCan my Frankenstein of a time series regression model — inspired by Prophet — compete with the real deal?Nov 25, 2023A response icon1Nov 25, 2023A response icon1
Published inTDS ArchiveFalse Prophet: a Homemade Time Series Regression ModelBorrowing ideas from Meta’s Prophet to build a powerful time series regression modelOct 31, 2023A response icon5Oct 31, 2023A response icon5