Sample Data Sources

[kennethshsu]

Where is the source of the sample data that comes pre-loaded by the package? Is it downloaded from somewhere?

I am seeing some numbers in xyz not reconciling exactly to the examples in the Fridland text.

xyz = cl.load_sample("xyz")

# Case Reserves
# vs Page 281 https://www.casact.org/sites/default/files/database/studynotes_friedland_estimating.pdf
xyz["Incurred"] - xyz["Paid"]

# Incremental Paid
# vs Page 281 https://www.casact.org/sites/default/files/database/studynotes_friedland_estimating.pdf
xyz["Paid"]

[jbogaardt]

https://github.com/casact/chainladder-python/tree/master/chainladder/utils/data

[kennethshsu]

I mean how or where did you download the original data set? Was any transformation performed before putting the csvs here?

[jbogaardt]

I believe I grabbed the data from the R chainladder package.

[kennethshsu]

Hmm, I don't think I see xyz here, any other guesses?

It's not a big deal, just a few numbers are off, so I was wondering if we did any clean up such as transforming or reformatting before we put them in the util folder.

[jbogaardt]

Hmm then idk. Can you correct it?