README.md

In selva86/car2: Performance analysis and companion functions such as AUROC, IV, WOE calculation for binary classification models (logistic, discriminant etc)

car2

Extends capabilities of 'car' to include companion functions for logistic regression

To install: The latest development version: library(devtools); install.github("car2", "selva86");

The following functions are covered in the package:

1. AUROC: Computes Area under ROC curve.
2. Concordance: Calculates the concordance and discordance statistic.
3. confusionMatrix: Compute the Good / Bad match matrix.
4. IV: Compute the information value for a given categorical variable.
5. kappaCohen: Computes the kappaCohen statistic, an indicator of the predictive power of the model.
6. misClassError: Computes the percentage misclassification error.
7. plotROC: Plots a beautiful ROC curve using the ggplot2 framework.
8. sensitivity: Computes the True Positive rate.
9. SimData: A data with binary response variable (1/0) and a categorical X variable, used to demo the calculation of WOE.
10. somersD: Computes somersD statistic which equals to Concordance - Discordance.
11. Specificity: Computes the specificity of the given Actuals (1/0's) and the predicted probability scores and given threshold. It representative of prediction coverage of non-events, i.e. what portion of the non-events were predicted correctly.
12. WOE: The weights of evidence of a given categorical X variable againsts as given binary response (1/0's).
13. WOETable: Compute the WOE table showing the WOE and IV values for each group in the categorical variable.
14. youdensIndex: Computes the youden's Index statistic.

selva86/car2 documentation built on May 29, 2019, 5:55 p.m.