Code Examples

Quick, focused examples demonstrating specific Causal Agent features and common use cases. These examples are designed to be copy-and-paste ready for your own projects.

Example Categories

Dataset Properties and Method Selection Gallery

Visual guide showing how different dataset characteristics lead to different method selections:

  • Decision tree navigation examples

  • Method exclusion explanations

  • Side-by-side comparisons

  • Interactive decision framework

Decision Path Comparisons

Side-by-side comparisons of how similar datasets with slight variations lead to different method selections:

  • Randomized vs. observational data

  • Cross-sectional vs. panel data

  • Strong vs. weak instruments

  • Good vs. poor covariate overlap

Basic Usage Examples

Simple, straightforward examples for common tasks:

  • Loading data and running basic analysis

  • Interpreting Causal Agent output

  • Saving and loading results

  • Basic visualization of results

Data Preparation Examples

Preparing your data for causal analysis:

  • Handling missing values

  • Creating treatment and control groups

  • Time series data preparation

  • Merging multiple datasets

Method-Specific Examples

Examples for each causal inference method:

  • RCT analysis with Causal Agent

  • Difference-in-Differences implementation

  • Instrumental Variables analysis

  • Regression Discontinuity Design

  • Propensity Score methods

Configuration Examples

Customizing Causal Agent for your needs:

  • LLM provider configuration

  • Custom method selection

  • Batch processing setup

  • Performance optimization

Integration Examples

Using Causal Agent with other tools and workflows:

  • Pandas DataFrame integration

  • Scikit-learn pipeline integration

  • Apache Spark for big data

  • MLflow for experiment tracking

Quick Start Examples

Basic Analysis

from causal_agent import CausalAgent
import pandas as pd

# Load your data
data = pd.read_csv('your_data.csv')

# Initialize Causal Agent
agent = CausalAgent()

# Run analysis
result = agent.analyze(
    data=data,
    treatment='treatment_column',
    outcome='outcome_column'
)

# View results
print(result.summary())

Method-Specific Analysis

from causal_agent.methods import DifferenceInDifferences

# Use specific method
did = DifferenceInDifferences()
result = did.estimate(
    data=data,
    treatment='policy_implemented',
    outcome='outcome_measure',
    time_var='year',
    unit_var='state'
)

Batch Processing

from causal_agent import CausalAgent

agent = CausalAgent()

# Process multiple datasets
datasets = ['data1.csv', 'data2.csv', 'data3.csv']
results = []

for dataset in datasets:
    data = pd.read_csv(dataset)
    result = agent.analyze(data, 'treatment', 'outcome')
    results.append(result)

Custom Configuration

from causal_agent import CausalAgent, Config

# Custom configuration
config = Config(
    llm_provider='openai',
    model='gpt-4',
    temperature=0.1,
    max_tokens=1000
)

agent = CausalAgent(config=config)

Example Datasets

All examples include sample datasets that you can use to practice:

  • Synthetic RCT Data: Simulated randomized experiment

  • Policy Evaluation Data: Government policy implementation

  • Marketing Campaign Data: Advertising effectiveness study

  • Medical Treatment Data: Healthcare intervention analysis

  • Education Program Data: Educational intervention evaluation

Usage Tips

  • Copy and Modify: All examples are designed to be easily adapted

  • Error Handling: Examples include proper error handling patterns

  • Best Practices: Follow the patterns shown for robust analysis

  • Documentation: Each example includes detailed comments

  • Testing: Examples include basic validation and testing approaches

Contributing Examples

We welcome community contributions of new examples! See our ../../development/contributing guide for:

  • Example submission guidelines

  • Code style requirements

  • Documentation standards

  • Review process