Example notebooks#

These examples are also available on GitHub. You can run them locally after cloning DoWhy and installing Jupyter. Or you can run them directly in a web browser using the Binder environment.

Introductory examples#

Basic Example for Calculating the Causal Effect
../_images/effect-estimation-estimand-expression.png
Level: Beginner
Task: Effect estimation
Basic Example for Graphical Causal Models
../_images/graph-xyz.png
Level: Beginner
Task: Intervention via GCM
Basic Example for generating samples from a GCM
../_images/draw-samples.png
Level: Beginner
Task: Sample generation
Confounding Example: Finding causal effects from observed data
../_images/confounder-example.png
Level: Beginner
Task: Effect estimation
Conditional Average Treatment Effects (CATE) with DoWhy and EconML
../_images/conditional-treatment-effect.png
Level: Beginner
Task: Conditional effect estimation
Tutorial on Causal Inference and its Connections to Machine Learning (Using DoWhy+EconML)
../_images/dowhy-econml-tutorial.png
Level: Beginner
Task: All
Do-sampler Introduction
Level: Beginner
Task: Interventions

Real world-inspired examples#

Exploring Causes of Hotel Booking Cancellations
../_images/hotel-bookings.png
Level: Beginner
Task: Effect estimation
Estimating the Effect of a Member Rewards Program
../_images/membership-program-graph.png
Level: Beginner
Task: Effect estimation
Causal Attributions and Root-Cause Analysis in an Online Shop
../_images/online-shop.png
Level: Beginner
Task: Quantify causal effects and root cause analysis via GCM
Finding the Root Cause of Elevated Latencies in a Microservice Architecture
../_images/microservice-architecture.png
Level: Beginner
Task: Root cause analysis, intervention, counterfactual via GCM
Impact of 401(k) eligibility on net financial assets
../_images/401k-example.png
Level: Advanced
Task: Intervention via GCM
Finding Root Causes of Changes in a Supply Chain
../_images/supply-chain1.png
Level: Advanced
Task: Root cause analysis via GCM
Decomposing the Gender Wage Gap
../_images/cps2015-shapley-values.png
Level: Advanced
Task: Causal Change Attribution via GCM
Estimating intrinsic causal influences in real-world examples
../_images/icc-example.jpg
Level: Advanced
Task: Intrinsic causal influence via GCM
Counterfactual Analysis in a Medical Case
../_images/gcm-counterfactual-example.png
Level: Advanced
Task: Counterfactuals via GCM
Falsification of User-Given Directed Acyclic Graphs
../_images/gcm-falsify-example.png
Level: Advanced
Task: Falsifying Causal Graph
Counterfactual Fairness
../_images/counterfactual_fairness_dowhy.png
Level: Advanced
Task: Fairness estimation via GCM
Causal attribution to sales growth and spend intervention
../_images/sales_attribution.png
Level: Advanced
Task: Causal Inference

Examples on benchmark datasets#

DoWhy example on ihdp (Infant Health and Development Program) dataset
Level: Advanced
Task: Effect inference
DoWhy example on the Lalonde dataset
Level: Advanced
Task: Effect inference
Applying refutation tests to the Lalonde and IHDP datasets
Level: Advanced
Task: Effect inference
Impact of 401(k) eligibility on net financial assets
Level: Advanced
Task: GCM inference
Demo for DoWhy Causal Prediction on MNIST
Level: Advanced
Task: Prediction
Lalonde Pandas API Example
Level: Advanced
Task: Do Sampler
Counterfactual Fairness
Level: Advanced
Task: Fairness estimation
DoWhy example on Twins dataset
Level: Advanced
Task: Effect inference

Modeling and refuting causal assumptions#

Different ways to load an input graph
Level: Beginner
Task: All
Causal Discovery example
Level: Beginner
Task: All
Falsification of User-Given Directed Acyclic Graphs
Level: Advanced
Task: All
Sensitivity Analysis for Regression Models
Level: Beginner
Task: Effect inference
Sensitivity analysis for non-parametric causal estimators
Level: Advanced
Task: Effect inference
Iterating over multiple refutation tests
Level: Beginner
Task: Effect inference
Assessing Support and Overlap with OverRule
Level: Advanced
Task: Effect inference
A Simple Example on Creating a Custom Refutation Using User-Defined Outcome Functions
Level: Advanced
Task: Effect inference
Ranking of estimation methods for a given dataset
Level: Advanced
Task: Effect inference

Miscellaneous#

DoWhy: Different estimation methods for causal inference
Level: Beginner
Task: Effect inference
Simple example on using Instrumental Variables method for estimation
Level: Beginner
Task: Effect inference
DoWhy: Interpreters for Causal Estimators
Level: Beginner
Task: Effect inference
Mediation analysis with DoWhy: Direct and Indirect Effects
Level: Advanced
Task: Effect inference
Estimating effect of multiple treatments
Level: Advanced
Task: Effect inference
Finding optimal adjustment sets
Level: Advanced
Task: Effect inference
Identifying Effect using ID Algorithm
Level: Advanced
Task: Effect inference
Example to demonstrate optimized backdoor variable search for Causal Identification
Level: Advanced
Task: Effect inference