---
title: Probabilistic Programming & Causal Inference
description: Bayesian modeling and MCMC/variational inference libraries, plus causal effect estimation and discovery.
tags: [data-science, machine-learning, probabilistic]
---

# Probabilistic Programming & Causal Inference

Fetch when the target builds Bayesian models, runs MCMC or variational inference, estimates causal effects, or discovers causal structure. Each entry is one line and a documentation entry point; fetch install commands and API details from the entry point, never from memory. No entry is a recommendation.

## Probabilistic Programming & Bayesian Inference

| Tool | One line | Docs |
|---|---|---|
| PyMC | Bayesian modeling in Python | <https://www.pymc.io/> |
| ArviZ | exploratory analysis of Bayesian inference results | <https://python.arviz.org/> |
| Stan | the Stan probabilistic language and samplers | <https://mc-stan.org/> |
| CmdStanPy | Python interface to CmdStan | <https://mc-stan.org/cmdstanpy/> |
| Pyro | deep probabilistic programming on PyTorch | <https://pyro.ai/> |
| NumPyro | Pyro's NumPy/JAX backend | <https://num.pyro.ai/> |
| TensorFlow Probability | probabilistic layers and inference on TensorFlow | <https://www.tensorflow.org/probability> |
| BlackJAX | composable samplers in JAX | <https://blackjax-devs.github.io/blackjax/> |

## Causal Inference

| Tool | One line | Docs |
|---|---|---|
| DoWhy | causal-assumption-explicit effect estimation | <https://www.pywhy.org/dowhy/> |
| causal-learn | causal discovery algorithms | <https://causal-learn.readthedocs.io/> |
| EconML | ML-based heterogeneous treatment effects | <https://github.com/py-why/EconML> |
| CausalML | uplift modeling and causal ML | <https://causalml.readthedocs.io/> |
| DoubleML | double/debiased machine learning | <https://docs.doubleml.org/> |
| Tigramite | causal discovery for time series | <https://github.com/jakobrunge/tigramite> |