book

Statistical Rethinking, 2nd Edition

by Richard McElreath

March 2020

Beginner to intermediate

612 pages

22h 10m

English

Chapman and Hall/CRC

Read now

Unlock full access

AudienceTeaching strategyHow to use this bookInstalling the rethinking R packageAcknowledgments
1.1. Statistical golems1.2. Statistical rethinking1.3. Tools for golem engineering1.4. Summary
2.1. The garden of forking data2.2. Building a model2.3. Components of the model2.4. Making the model go2.5. Summary2.6. Practice

3.1. Sampling from a grid-approximate posterior3.2. Sampling to summarize3.3. Sampling to simulate prediction3.4. Summary3.5. Practice
4.1. Why normal distributions are normal4.2. A language for describing models4.3. Gaussian model of height4.4. Linear prediction4.5. Curves from lines4.6. Summary4.7. Practice
5.1. Spurious association5.2. Masked relationship5.3. Categorical variables5.4. Summary5.5. Practice
6.1. Multicollinearity6.2. Post-treatment bias6.3. Collider bias6.4. Confronting confounding6.5. Summary6.6. Practice
7.1. The problem with parameters7.2. Entropy and accuracy7.3. Golem taming: regularization7.4. Predicting predictive accuracy7.5. Model comparison7.6. Summary7.7. Practice
8.1. Building an interaction8.2. Symmetry of interactions8.3. Continuous interactions8.4. Summary8.5. Practice
9.1. Good King Markov and his island kingdom9.2. Metropolis algorithms9.3. Hamiltonian Monte Carlo9.4. Easy HMC: ulam9.5. Care and feeding of your Markov chain9.6. Summary9.7. Practice
10.1. Maximum entropy10.2. Generalized linear models10.3. Maximum entropy priors10.4. Summary
11.1. Binomial regression11.2. Poisson regression11.3. Multinomial and categorical models11.4. Summary11.5. Practice
12.1. Over-dispersed counts12.2. Zero-inflated outcomes12.3. Ordered categorical outcomes12.4. Ordered categorical predictors12.5. Summary12.6. Practice
13.1. Example: Multilevel tadpoles13.2. Varying effects and the underfitting/overfitting trade-off13.3. More than one type of cluster13.4. Divergent transitions and non-centered priors13.5. Multilevel posterior predictions13.6. Summary13.7. Practice
14.1. Varying slopes by construction14.2. Advanced varying slopes14.3. Instruments and causal designs14.4. Social relations as correlated varying effects14.5. Continuous categories and the Gaussian process14.6. Summary14.7. Practice
15.1. Measurement error15.2. Missing data15.3. Categorical errors and discrete absences15.4. Summary15.5. Practice
16.1. Geometric people16.2. Hidden minds and observed behavior16.3. Ordinary differential nut cracking16.4. Population dynamics16.5. Summary16.6. Practice

Overview

Statistical Rethinking: A Bayesian Course with Examples in R and Stan, Second Edition builds knowledge/confidence in statistical modeling. Pushes readers to perform step-by-step calculations (usually automated.) Unique, computational approach.