AUC: functionals of ROC curve In ROC: utilities for ROC, with microarray focus

Description

various functionals of ROC (Receiver Operating Characteristic) curves

Usage

 ```1 2 3 4``` ```AUC(rocobj) AUCi(rocobj) pAUC(rocobj,t0) pAUCi(rocobj,t0) ```

Arguments

 `rocobj` element of class rocc `t0` FPR point at which TPR is evaluated or limit in (0,1) to integrate to

Details

AUC, pAUC, AUCi and pAUCi compute the Area Under the Curve.

AUC and pAUC employ the trapezoidal rule. AUCi and pAUCi use integrate().

AUC and AUCi compute the area under the curve from 0 to 1 on the x-axis (i.e., the 1 - specificity axis).

pAUC and pAUCi compute the are under the curve from 0 to argument t0 on the x-axis (i.e., the 1 - specificity axis).

Elements of class rocc can be created by rocdemo.sca() or other constructors you might make using the code of rocdemo.sca() as a template.

Author(s)

Vince Carey (stvjc@channing.harvard.edu)

References

Rosner, B., 2000, Fundamentals of Biostatistics, 5th Ed., pp. 63–65

Duda, R. O., Hart, P. E., Stork, D. G., 2001 Pattern Classification, 2nd Ed., p. 49

rocdemo.sca

Examples

 ```1 2 3 4 5 6 7``` ```set.seed(123) R1 <- rocdemo.sca( rbinom(40,1,.3), rnorm(40), dxrule.sca, caseLabel="new case", markerLabel="demo Marker" ) print(AUC(R1)) print(pAUC(R1,.3)) print(pAUCi(R1,.3)) print(ROC(R1,.3)) ```

Example output

```NA in cutpts forces recomputation using smallest gap
[1] 0.456044
[1] 0.04010989
[1] 0.04236287
[1] 0.2
```

ROC documentation built on Nov. 8, 2020, 5:23 p.m.