So you think you know about linear regression ...
summary
Explains how linear regression is a Bayesian generative model, where ridge regression = Gaussian prior and multicollinearity comes from a smeared likelihood not magic; shows tradeoffs between MCMC gives full uncertainty and MAP / SciPy / scikit‑learn give fast point estimates; runs concrete examples — fantasy sports, heavy‑equipment pricing, and a once‑profitable HFT strategy that later fails — and demos when changing error models or priors fixes instability.
AI-written summary · about
chapters (3)
related talks
39mthe idea behind Automatic Relevance Determination and Bayesian Interpolation
01h 26mProbabilistic Programming and Bayesian Computing with PyMC
01h 30mA Primer on Gaussian Processes for Regression Analysis
48mTBA
31mLogistic Regression Behind the Scenes
14mProbabilistic Programming in Python
31mBut what is a Gaussian process? Regression while knowing how certain you are
42mBayesian Non-parametric Models for Data Science using PyMC3