Analytics Store Tutorials - Insurance Data Analysis

A comprehensive tutorial series demonstrating the capabilities of the analytics_store package using insurance industry datasets. These exercises cover data loading, model validation, performance monitoring, and advanced analytics techniques.

📚 Tutorial Overview

This repository contains hands-on exercises organized by difficulty level and topic:

Getting Started (Start Here!)

0. Project Setup and Polars Basics - Environment setup and data manipulation fundamentals

Beginner Tutorials

1. Introduction to Lift Analysis - Understanding lift curves with insurance claim predictions
2. ROC Curve Analysis - Evaluating binary classification models for fraud detection
3. Regression Metrics - Analyzing premium prediction models

Intermediate Tutorials

4. Model Comparison - Comparing multiple models using double lift analysis
5. Data Loading from SQL - Connecting to databases and loading insurance data
6. Population Testing - Statistical tests for comparing customer segments

Advanced Tutorials

7. Model Monitoring & Drift Detection - Tracking model performance over time
8. Snowflake Integration - Working with cloud data warehouses
9. End-to-End Pipeline - Complete workflow from data loading to reporting

🎯 Learning Objectives

By completing these tutorials, you will learn to:

• Use analytics_store for comprehensive model evaluation
• Work with Polars DataFrames for high-performance data analysis
• Load data from various sources (CSV, SQL, Snowflake)
• Evaluate classification and regression models
• Monitor model performance and detect data drift
• Compare multiple models and scoring approaches
• Generate professional analytics reports

📋 Prerequisites

• Python 3.8 or higher
• Basic understanding of Python and pandas/polars
• Familiarity with machine learning concepts (helpful but not required)

🚀 Getting Started

Installation

1. Clone this repository:

git clone https://github.com/Wicks-Analytics/analytics_tutorials.git
cd analytics_tutorials

2. Choose your environment manager:

Option A: Using uv (Recommended - Fast & Modern)

uv is an extremely fast Python package installer and resolver.

# Install uv (if not already installed)
# Windows (PowerShell)
powershell -c "irm https://astral.sh/uv/install.ps1 | iex"

# macOS/Linux
curl -LsSf https://astral.sh/uv/install.sh | sh

# Create virtual environment and install dependencies
uv venv
source .venv/bin/activate  # On Windows: .venv\Scripts\activate
uv pip install -e .

# Install optional dependencies
uv pip install -e ".[jupyter]"  # For Jupyter notebooks
uv pip install -e ".[snowflake]"  # For Snowflake tutorials
uv pip install -e ".[all]"  # Install everything

# Install analytics_store
uv pip install git+https://github.com/Wicks-Analytics/analytics_store

Option B: Using pip (Traditional)

# Create a virtual environment
python -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate

# Install dependencies
pip install -r requirements.txt

# Install analytics_store
pip install git+https://github.com/Wicks-Analytics/analytics_store

Running Tutorials

Option 1: Python Scripts

Start with Tutorial 00, then proceed through the beginner tutorials:

# Start here - setup and Polars basics
python tutorials/00_getting_started/00_setup_and_polars_basics.py

# Then proceed to beginner tutorials
python tutorials/01_beginner/01_lift_analysis.py
python tutorials/01_beginner/02_roc_analysis.py
python tutorials/01_beginner/03_regression_metrics.py

Option 2: Jupyter Notebooks (Recommended for Learning)

Interactive notebook versions are available in the notebooks/ folder:

# Install Jupyter
pip install jupyter jupyterlab

# Start JupyterLab
jupyter lab

# Navigate to notebooks/01_beginner/01_lift_analysis.ipynb

Converting Scripts to Notebooks

Generate notebook versions from Python scripts:

python convert_to_notebooks.py

📁 Repository Structure

analytics_tutorials/
├── README.md                          # This file
├── QUICKSTART.md                      # 5-minute setup guide
├── TUTORIAL_INDEX.md                  # Complete tutorial reference
├── requirements.txt                   # Python dependencies
├── setup_database.py                  # Script to set up local SQLite database
├── convert_to_notebooks.py            # Convert Python scripts to Jupyter notebooks
├── data/                              # Sample datasets
│   ├── insurance_claims.csv          # Dummy insurance claims data
│   ├── insurance_policies.csv        # Dummy policy data
│   └── fraud_predictions.csv         # Model predictions for fraud detection
├── tutorials/                         # Tutorial exercises (Python scripts)
│   ├── 00_getting_started/
│   │   └── 00_setup_and_polars_basics.py
│   ├── 01_beginner/
│   │   ├── 01_lift_analysis.py
│   │   ├── 02_roc_analysis.py
│   │   └── 03_regression_metrics.py
│   ├── 02_intermediate/
│   │   ├── 04_model_comparison.py
│   │   ├── 05_sql_integration.py
│   │   └── 06_population_testing.py
│   └── 03_advanced/
│       ├── 07_model_monitoring.py
│       ├── 08_snowflake_integration.py
│       └── 09_end_to_end_pipeline.py
├── notebooks/                         # Jupyter notebook versions
│   ├── README.md                      # Notebook usage guide
│   ├── 01_beginner/
│   ├── 02_intermediate/
│   └── 03_advanced/
├── solutions/                         # Solutions with detailed explanations
│   ├── 01_beginner/
│   ├── 02_intermediate/
│   └── 03_advanced/
├── utils/                             # Helper utilities
│   ├── __init__.py
│   ├── data_generators.py            # Functions to generate dummy data
│   └── database_helpers.py           # Database connection utilities
└── outputs/                           # Generated plots and reports (gitignored)

📊 Sample Datasets

The repository includes realistic dummy insurance datasets:

• insurance_claims.csv - Historical claims data with features and outcomes
• insurance_policies.csv - Policy information and customer demographics
• fraud_predictions.csv - Model predictions for fraud detection exercises

All data is synthetically generated and does not contain real customer information.

🔌 Database Integration

SQLite (Local)

Run the setup script to create a local SQLite database:

python setup_database.py

PostgreSQL/MySQL

Update the connection strings in utils/database_helpers.py with your credentials.

Snowflake

For Snowflake tutorials, you'll need:

• Snowflake account credentials
• Appropriate permissions to read data
• Update config/snowflake_config.py with your connection details

📝 Tutorial Descriptions

Beginner Level

01 - Lift Analysis Learn to calculate and visualize lift curves for insurance claim predictions. Understand how to interpret lift metrics and identify model performance across deciles.

02 - ROC Analysis Evaluate binary classification models using ROC curves and AUC scores. Learn to find optimal thresholds for fraud detection models.

03 - Regression Metrics Analyze regression model performance for premium predictions. Calculate RMSE, MAE, R-squared, and create diagnostic plots.

Intermediate Level

04 - Model Comparison Compare multiple scoring models using double lift analysis. Understand joint and conditional lift metrics.

05 - SQL Integration Load insurance data from SQL databases using Polars. Learn efficient data loading patterns and query optimization.

06 - Population Testing Perform statistical tests to compare different customer segments. Use t-tests and Mann-Whitney U tests with effect size calculations.

Advanced Level

07 - Model Monitoring Implement comprehensive model monitoring with drift detection. Track feature drift using PSI and performance metrics over time.

08 - Snowflake Integration Connect to Snowflake data warehouse and perform analytics at scale. Learn best practices for cloud data integration.

09 - End-to-End Pipeline Build a complete analytics pipeline from data loading through model evaluation to automated reporting.

🎓 Learning Path

Recommended order for beginners:

1. Start with tutorials 01-03 to understand core concepts
2. Move to tutorials 04-06 for practical integration skills
3. Complete tutorials 07-09 for production-ready workflows

For experienced users:

• Jump directly to intermediate or advanced tutorials
• Use beginner tutorials as reference material

� Testing & Development

Running Tests

This project uses pytest for testing:

# Run all tests
pytest

# Run with coverage
pytest --cov=. --cov-report=term-missing

# Run tests in parallel (faster)
pytest -n auto

Code Quality

We use Ruff and Black for code quality:

# Lint with Ruff
ruff check .

# Format with Ruff
ruff format .

# Format with Black
black .

Pre-commit Hooks

Install pre-commit hooks to automatically check code quality:

uv pip install -e ".[dev]"
pre-commit install

Continuous Integration

All pull requests are automatically tested with GitHub Actions:

• ✅ Linting (Ruff)
• ✅ Formatting (Black + Ruff)
• ✅ Tests (Python 3.8-3.12, Ubuntu/Windows/macOS)
• ✅ Tutorial validation

�� Contributing

We welcome contributions! If you have ideas for new tutorials or improvements:

1. Fork the repository
2. Create a feature branch
3. Follow the guidelines in CONTRIBUTING.md
4. Submit a pull request

See CONTRIBUTING.md for detailed development setup and guidelines.

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.

🙏 Acknowledgments

• Built using the analytics_store package by Wicks Analytics LTD
• Powered by Polars for high-performance data processing

📧 Support

For questions or issues:

• Open an issue on GitHub
• Check the analytics_store documentation
• Review the solutions folder for detailed explanations

---

Happy Learning! 🎉