Modeling Extreme Events with PyMC
summary
Models New York extreme rainfall with PyMC using annual maxima (about 50 annual maxima) and fits an extreme value distribution to estimate return levels; the single-day record of 8.4 inches maps to roughly a once in 83 years event under the baseline model. A Gaussian process prior that borrows strength from nearby stations (scaled experiments up to 3,500 stations) reduces posterior uncertainty roughly by half, sampling takes minutes, and posterior return-level curves and credible bands are produced.
AI-written summary · about
chapters (6)
- 0:00Introduction
- 1:09Extreme events statistics
- 6:49Bayesian modeling basics
- 11:03NY rainfall modeling
- 17:51Gaussian process enhancement
- 25:18Key takeaways and Q&A session
related talks
01h 27mBayesian Time Series
01h 27mProbabilistic Python: An Introduction to Bayesian Modeling with PyMC
42mBayesian Non-parametric Models for Data Science using PyMC3
44mAn introduction to PyMC3
36mClusterf*ck: A Practical Guide to Bayesian Hierarchical Modeling in PyMC3
01h 26mProbabilistic Programming and Bayesian Computing with PyMC
42mSolving Real-World Business Problems with Bayesian Modeling
34mFast, beautiful and easy Bayesian modelling: Can you have it all?