auc.test: AUC based independence test
In liyistat/knnAUC: k-nearest neighbors AUC test

Description Usage Arguments Value Author(s) References

AUC based independence test. We evaluate the significance of the AUC based statistic (Null hypothesis: AUC = 0.5, that is to say, X have no prediction ability for Y).

1	auc.test(testy, yhat)

`testy`	A binary (0 or 1) vector, which will be predicted.
`yhat`	A vector, prediction probabilities of testy equals to 1.

result A list of average AUC estimator and pvalue of AUC based significance test.

Yi Li, liyistat@gmail.com

Yi Li, Xiaoyu Liu, et al. knnAUC : k-nearest neighbors AUC test.BMC Bioinformatics.2018.

Mason, S. J. and Graham, N. E. (2002), Areas beneath the relative operating characteristics (ROC) and relative operating levels (ROL) curves: Statistical significance and interpretation. Q.J.R. Meteorol. Soc. doi:10.1256/003590002320603584

liyistat/knnAUC documentation built on May 12, 2019, 10:51 a.m.