Open ColtAllen opened 2 years ago
It seems the summary_data_from_transaction_data
function is worth looking into as well. Any outstanding issues involving monetary value need to be resolved before GammaGammaModel
is added.
After reviewing this paper, the monetary value calculations are being done properly:
http://brucehardie.com/notes/022/RFM_summary_in_Excel.pdf
However, a test should still be added to ensure the output of calibration_and_holdout_data
matches that of the paper.
This is an unresolved issue with the legacy Lifetimes library that I think warrants further investigation. If the
calibration_and_holdout_data
function inutils.py
is indeed not calculating monetary value correctly between the holdout (training) and calibration (test) datasets, then this is a serious bug that will impact the ability to properly evaluate theGammaGamma
model.