EMP (version 1.0.2)

EMP-package: Expected Maximum Profit for Credit Scoring.

Description

The EMP measure is an alternative to AUC that includes the expected profit of a given model, when compared to a baseline (no model used). Presented in Verbraken et al. (2014) as a preferred measure for credit scoring in any profit-driven environment. This implementation assumes an LGD distribution with two point masses, and a constant ROI.

Details

ll{ Package: EMP Type: Package Version: 1.0.2 Date: 2014-08-24 License: GPL (>=3) } The package consists of only two functions, EMP_CS and calcROC. The second one estimates the convex hull of any ROC function starting from the scores and the classes, while the second estimates the EMP in this particular case. See the help pages of each function for details.

References

Verbraken, T., Bravo, C., Weber, R. and Baesens, B. (2014). Development and application of consumer credit scoring models using profit-based classification measures. European Journal of Operational Research. 238 (2): 505 - 513. Available Online: http://www.sciencedirect.com/science/article/pii/S0377221714003105

Examples

# Dummy dataset.
score.ex <- runif(1000, 0, 1)
class.ex <- lapply(score.ex, function(x){rbinom(1,1,x)})
class.ex <- unlist(class.ex)

# Calculate ROC curve.
ROC.ex <- calcROC(score.ex, class.ex)

# Estimates EMP with point masses in 0.1 and 0.9, and 0.1 ROI
EMP.ex <- EMP_CS(ROC.ex, 0.1, 0.1, 0.1)
EMP.ex