PinnedPublished inTDS ArchiveWhen Is Bayesian Machine Learning Actually Useful?Personal thoughts about a somewhat controversial paradigmJan 27, 20223Jan 27, 20223
Published inTDS ArchiveWinning with Simple, not even Linear Time-Series ModelsIf your dataset is small, the subsequent ideas might be usefulMay 10, 20231May 10, 20231
Published inDataDrivenInvestorVarying Coefficient GARCHLet’s make GARCH have varying coefficients to handle non-linear conditional variance.Jan 19, 20231Jan 19, 20231
Published inTDS ArchiveWhen Point Forecasts Are Completely UselessWhile point forecasts are very popular, be aware of some unlucky pitfallsJan 1, 2023Jan 1, 2023
Published inTDS ArchiveWhy I prefer probabilistic forecasts — hitting time probabilitiesPoint forecasts are good for making decisions. With probabilistic forecasts, you can also make the right ones.Dec 6, 2022Dec 6, 2022
Published inDataDrivenInvestorRandom Forests and Boosting for ARCH-like volatility forecastsTree models are not just useful for point and mean forecasts.Oct 7, 20222Oct 7, 20222
Published inTDS ArchiveForecasting with Decision Trees and Random ForestsRandom Forests are flexible and powerful when it comes to tabular data. Do they also work for time-series forecasting? Let’s find out.Sep 19, 20223Sep 19, 20223
Published inDataDrivenInvestorMultivariate GARCH with Python and TensorflowOne primary limitation of GARCH is the restriction to a single dimensional time-series. In reality, however, we are typically dealing with…Sep 11, 20221Sep 11, 20221
Published inTDS ArchiveCointegrated time-series and when differencing might be badYou have heard about integrated time-series data but what about cointegration?Aug 25, 2022Aug 25, 2022
Published inTDS ArchiveFacebook Prophet, Covid and why I don’t trust the ProphetFacebook Prophet is highly popular for time-series forecasting. Let me show you why I am not a big fan and what else you can use.Aug 8, 20223Aug 8, 20223