Utilities

Utilities contains example datasets and extra functionality to facilitate a reserving workflow.

Sample Datasets

A variety of datasets can be loaded using :func:load_sample(). These are sample datasets that are used in a variety of examples within this documentation.

Dataset	Description
abc	ABC Data
auto	Auto Data
berqsherm	Data from the Berquist Sherman paper
cc_sample	Sample Insurance Data for Cape Cod Method in Struhuss
clrd	CAS Loss Reserving Database
genins	General Insurance Data used in Clark
ia_sample	Sample data for Incremental Additive Method in Schmidt
liab	more data
m3ir5	more data
mcl	Sample insurance data for Munich Adjustment in Quarg
mortgage	more data
mw2008	more data
mw2014	more data
quarterly	Sample data to demonstrate changing Triangle grain
raa	Sample data used in Mack Chainladder
ukmotor	more data
usaa	more data
usauto	more data

Chainladder Persistence

All estimators can be persisted to disk or database using to_json or to_pickle. Restoring the estimator is as simple as cl.read_json or cl.read_pickle.

import chainladder as cl
model_json = cl.Chainladder().fit(cl.load_sample('raa')).to_json()
model_json

'{"params": {}, "__class__": "Chainladder"}'

cl.read_json(model_json)

Chainladder()

In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook.
On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.

The saved Estimator does not retain any fitted attributes, nor does it retain the data on which it was fit. It is simply the model definition. However, the Triangle itself can also be saved allowing for a full rehydration of the original model.

# Dumping triangle to JSON
triangle_json = cl.load_sample('raa').to_json()

# Recalling model and Triangle and rehydrating the results
cl.read_json(model_json).fit(cl.read_json(triangle_json)).ibnr_.sum('origin')

/home/docs/checkouts/readthedocs.org/user_builds/chainladder-python/conda/latest/lib/python3.11/site-packages/chainladder/utils/utility_functions.py:113: FutureWarning: Passing literal json to 'read_json' is deprecated and will be removed in a future version. To read from a literal string, wrap it in a 'StringIO' object.
  y = pd.read_json(j["data"], orient="split", date_unit="ns")

52135.228261210155

Warning

Some features of estimators may not be json-serializable, such as a virtual_column or a callable hyperparameter. In these cases, JSON serialization will fail.