09-25, 12:05–12:40 (Europe/Amsterdam), Nebula
Forecasting time series can be messy, data is often missing, noisy, or full of structural changes like holidays, outliers, or evolving patterns. This talk shows how to build interpretable time series decomposition models using PyMC, a modern probabilistic programming library.
We’ll break time series into trend, seasonality, and noise components using engineered time features (e.g., Fourier and Radial Basis Functions). You’ll also learn how to model correlated series using hierarchical priors, letting multiple time series "learn from each other." As a case study, we’ll analyze Formula 1 lap time data to compare drivers and explore performance consistency using Bayesian posteriors.
This is a hands-on, code-first talk for data scientists, ML engineers, and researchers curious about Bayesian modeling (or Formula 1). Familiarity with Python and basic statistics is helpful, but no deep knowledge of Bayes is required.
We’ll build decomposition models from scratch using PyMC, focusing on:
* 0–5 min: Introducing time series decomposition and when you might need it
* 5–10 min: Building a baseline model (trend + noise)
* 10–15 min: Adding seasonal components and outlier indicators
* 15–20 min: Modeling correlated time series with hierarchical priors
* 20–25 min: Using posteriors for forecasting and anomaly detection
* 25–30 min: Wrap up and comparison with ARIMA/Prophet
This session prioritizes practical modeling over theory. Expect live coding on F1 data and takeaways ready to apply to your own time series projects.