CLAR

src.meliora.core.clar(df, predicted_ratings, realised_outcomes)

CLAR serves as a measure of ranking ability against LGD risk The cumulative LGD accuracy ratio (CLAR) curve can be treated as the equivalent of the Cumulative Accuracy Profile (CAP) curve. This test compares the cumulative percentage of correctly assigned realized LGD and the cumulative rate of observations in the predicted LGD bands. :param predicted_ratings: predicted LGD, can be ordinal or continuous :type predicted_ratings: pandas Series :param realised_outcomes: realised LGD, can be ordinal or continuous :type realised_outcomes: pandas Series

Returns

clar – Cumulative LGD Accuracy Ratio

Return type

scalar

References

[1] Ozdemir, B., Miu, P., 2009. Basel II Implementation. A Guide to Developing and Validating a Compliant Internal Risk Rating System. McGraw-Hill, USA. [2] See also: https://rdrr.io/cran/VUROCS/man/clar.html

Examples

>>> res = clar(predicted_ratings, realised_outcomes)
>>> print(res)