Project Overview

trikit is a collection of Actuarial reserving utilities written in Python. The library is built around triangle objects, which are data structures that organize losses by year of origin as a function of development period. trikit’s IncrTriangle and CumTriangle classes inherit from Pandas DataFrame, so triangle objects can be manipulated using the familiar Pandas API.

trikit is distributed with a number of sample datasets, in addition to the CAS Loss Reserving Database. Examples working with included sample datasets can be found in the Quickstart Guide, datasets reference and the CAS Loss Reserving Database reference.