About PyProtolinc

An Actuarial Projection Tool for Life Insurance Cash Flows.

This package allows to project future cash flows for portfolios of life & health insurance policies. It comes with a number of built-in standard products but can also be used to project custom products by the user.

Documentation

For extended documentation cf. https://pyprotolinc.readthedocs.io/en/latest/index.html.

Project Objectives

The key objective for PyProtolinc is to model cash flows for a variety of simple life and health insurance products, going forward also beyond stylized textbook examples.

The tool should provide a command line interface which can be used with configuration files as well as an extensible programming API which provides flexibility to adapt to own purposes.

Calculations should be laid out to deal with portfolios of insureds in a batch style and an attempt shall be made that forecast projections for reasonably large portfolios (of, say, a few 10s or 100s of thousands of policies) can be made in an acceptable amount of time (seconds or up to a few minutes rather than hours).

Basic Usage

Installation

To install from PyPI run:

pip install -U pyprotolinc

Alternatively, or for delevoplement clone (or download) the repository from https://github.com/mseehafer/PyProtolinc.git and run:

pip install -e .

from the root directory of the repository.

Quickstart

Usage is illustrated in detail by the prepared use cases in the examples folder. To try those out cd into the respective subfolder and run the tool from the command line:

pyprotolinc run

This will pick up the configuration file (config.yml) in the working directory (which points to the portfolio file in the subdirectory portfolio) and initiate a projection run. Once completed the (aggregate) results of the computation are written into a CSV file in the subfolder results. Now one can start playing around by changing the configuration. Note that the examples are commented in the documentation, cf. https://pyprotolinc.readthedocs.io/en/latest/examples/intro.html .