Python Libraries for Actuaries

I was recently trying to gather a list of sources that can used in Python for Actuarial Modelling purpose. Along with the available libraries and their capabilities that we can leverage for modelling, open source nature of python enables anyone to build their model from scratch. Below is the list of most commonly used python libraries in actuarial space.

List of Python libraries which can be used to for Actuarial models:

1. Lifelib - Python package with pre-built actuarial functions and examples.
2. Cashflower - Open Source package for Actuarial cashflow models.
3. SciPy - Scipy can be used for modeling, simulation, mathematical problems and more.
4. NumPy - Arrays, computations and much more can be done using NumPy. ?
5. Pandas - Data manipulation, Data handling, Data organization and Data analysis along with other useful functionalities.
6. Matplotlib and Plotly - Data presentation, Data plotting, static, aimated and interactive visualizations. ?
7. Numba - Speeding up the array calcualtions.
8. Bokeh - Powerful interactive visualizations based on data.
9. Scikit-Learn - Predictive data analytics in Python.
10. Tensorflow/Pytorch - Machine learning models.
11. Pyspark/Dask/SQLite - Big data storage and real time calculation using distributive methods.

Areas where Actuaries are moving towards the use of Python:

• Cash flow modeling (Liabilities/Assets/Reinvestment Projections)
• Valuation models
• Own Risk and Solvency Assememnts
• Reinsurance Projections and calculations
• Experience Studies
• Data checks/Model Checks/ Other model validations
• Machine learning/ Predictive analyses.
• Traditional Reporting/ Interative Reporting.
• Data compression techniques.