Examples gallery

Linear regression models

Multiple linear regression
Personality dimensions

Regression splines
Cherry blossoms

Hierarchical linear regression
Pigs growth

Hierarchical linear regression
Sleep deprivation

Hierarchical linear regression
Radon contamination

Bayesian workflow
Strack RRR analysis replication

Bayesian workflow
Police officer's dilemma

Fixed, Random Effects and Mundlak Machines
Reproducing McElreath's lecture on group-level confounding and intro to social networks

Robust linear regression
With the Student's T distribution

Predict new groups
With Hierarchical models

Polynomial regression
Learning gravity with Bayesian stats

Generalized Linear Models

Logistic regression
Model vote intention

Logistic regression
Model comparison with ArviZ

Hierarchical logistic regression
With the Binomial family

Regression for binary responses
Alternative link functions

Wald and Gamma regression
Australian insurance claims

Negative Binomial regression
Students absence

Count Regression with Variable Exposure
Accounting for varying exposure using offsets

Beta regression
With many datasets

Categorical outcomes
When the outcome is not a number

Circular regression
For directional statistics

Quantile regression
Model a percentile

MrP (Multilevel Regression and Post-stratification)
Survey Data and Representative Sampling

Zero inflated models
When the outcome is mostly zeros and or is overdispersed

Ordinal regression
Model ordered category outcomes

More advanced models

Distributional models
Many parameters simultaneously

Gaussian processes
In one dimension

Gaussian processes
In multiple dimensions

Survival modeling
Cat adoption times from Statistical Rethinking

Survival models in discrete time
An end to end review and application

Survival models in continuous time
From foundations to competing hazards

Orthogonal Polynomial regression
Polynomials that avoid multicollinearity

Non-centered parameterization
Per-prior and per-component control

Tools to interpret model outputs

Predictions
Compute and plot predictions on new data

Comparisons
Compare outputs between groups

Slopes
Determine how the response changes

Sensitivity analysis

Prior and likelihood sensitivity analysis
Evaluate the impact of priors and likelihoods on model outputs

Variable selection

Variable selection with Kulprit
Kullback-Leibler projections for Bayesian model selection

Alternative sampling backends

Using other samplers
JAX based samplers