Allen B. Downey: Books, Notebooks, and Workshops

Online versions of my books, their companion notebooks, and materials from my workshops. For my blog, newsletter, and everything else, see allendowney.com.

Books

Probably Overthinking It: how to use data to answer questions, avoid statistical traps, and make better decisions
You Would Choose Now: decades of survey data on changing attitudes in the United States
Think Stats, 3rd edition: exploratory data analysis in Python
Think Bayes, 2nd edition: Bayesian statistics in Python
Think Python, 3rd edition: an introduction to programming
Elements of Data Science: data science in Python for people with no programming experience
Modeling and Simulation in Python
Think Complexity, 2nd edition: complexity science and computational modeling
Think DSP: digital signal processing in Python
Data Structures and Information Retrieval in Python
The Little Book of Semaphores

Workshops and tutorials

Materials from workshops I have given at PyData, PyCon, ODSC, and elsewhere. Each one has notebooks you can work through on your own. Workshops marked self-study are not offered live now, but the material is still useful.

Bayesian statistics

Where to start: if you are new to probability, start with Bite Size Bayes. If you know Python and want the core ideas quickly, start with Bayesian Statistics Made Simple. Then move on to the PyMC workshops for real-world models. The book that goes with all of them is Think Bayes.

Bite Size Bayes: an introduction to Bayesian statistics in short notebooks, with no previous knowledge of probability assumed
Bayesian Statistics Made Simple: Bayesian methods using grid algorithms, for people who know Python (PyCon 2012–2016, SciPy 2019) · video
Bayesian Decision Analysis with PyMC: Beyond A/B Testing: A/B testing, Thompson sampling, and hierarchical models with PyMC (PyData Global 2025) · video
Bayesian Regression for Survey Data with PyMC: logistic, categorical, and ordered logistic regression (ODSC East 2025)
Bayesian Decision Analysis: from Bayes's theorem to Bayesian bandits, with NumPy and Pandas (ODSC East and PyData Global 2022) · video self-study
Bayesian Methods for Clinical Trials: PyMC examples and the FDA draft guidance on Bayesian methods self-study

Statistics and data analysis

Survey Data with Pandas: loading, validating, exploring, and modeling survey data with Pandas and StatsModels (PyCon US 2025)
Computational Statistics: effect sizes, sampling distributions, and hypothesis testing with resampling and simulation (PyCon 2015–2016, SciPy 2015, 2017) · video
Survival Analysis in Python: survival and hazard functions, computation first (PyData Global 2021) · video
Time Series Analysis with StatsModels: a chapter of Think Stats (PyData Global 2024) · video

Case studies

Each works on its own, and both are part of Elements of Data Science.

Political Alignment: fifty years of General Social Survey data on political alignment, outlook, and beliefs (PyData Global 2020) · videos: 1, 2, 3, 4
Recidivism: classification and algorithmic fairness, using the COMPAS risk scores

Other projects

Data Q&A: questions and answers about data and statistics