Welcome to the Geospatial A/B Testing Analysis repository! This project demonstrates a robust workflow for analyzing A/B test results in marketing campaigns with geospatial data, using Python.
In this project, we:
- Analyze geospatial data from an A/B marketing campaign.
- Validate assumptions using statistical tests (Shapiro-Wilk, Levene).
- Apply statistical methods (T-test, Mann-Whitney U) to assess conversion rates.
- Visualize results with insightful plots.
ab_test_analysis-2.ipynb: Notebook containing the detailed analysis workflow.dataset.csv: Dataset used for the analysis.
- Python Libraries:
pandas,numpyfor data manipulation.matplotlib,seabornfor data visualization.scipyfor statistical tests.
- Notebook Features:
- Markdown for documentation.
- Inline comments for clear explanations.
| Column | Description |
|---|---|
user_id |
Unique user identifier |
group |
Experiment group (A or B) |
latitude |
User's geographic latitude |
longitude |
User's geographic longitude |
conversion |
Conversion indicator (1=Yes, 0=No) |
-
Clone the repository:
git clone https://github.com/Anello92/Geospatial-AB-Testing.git
-
Navigate to the project directory:
cd Geospatial-AB-Testing -
Install the required Python libraries:
pip install -r requirements.txt
-
Open the notebook:
jupyter notebook ab_test_analysis-2.ipynb
- Statistical Insights:
- Normality and homogeneity tests.
- Significant differences in conversion rates between groups.
- Visualizations:
- Geospatial distributions of conversions.
- Comparative plots for group performance.
This project highlights the importance of:
- Rigorous statistical validation for A/B testing.
- Clear communication of findings using visuals.
- Handling and analyzing geospatial data effectively.
Contributions are welcome! If you have suggestions or improvements, feel free to open an issue or submit a pull request.
Thanks to all contributors and the open-source community for their invaluable tools and resources.
Made with ❤️.