R: Repeated cross validation of the AUC-RF process.
AUCRFcv
R Documentation
Repeated cross validation of the AUC-RF process.
Description
Performes a repeated cross validation analysis and computes the probability of selection for
each variable.
Usage
AUCRFcv(x, nCV = 5, M = 20)
Arguments
x
an object of class AUCRF.
nCV
number of folds in cross validation. By default a 5-fold cross validation is performed.
M
number of cross validation repetitions.
Details
The results of this repeated cross validation analysis are (1) a corrected estimation
of the predictive accuracy of the selected variables and (2) an estimate of the probability of selection for
each variable.
The AUC-RF algorithm is exhaustively described in Calle et. al.(2011).
Value
The same AUCRF object passed (see AUCRF) as argument but updated with the following
components:
cvAUC
mean of AUC values in test datasets of the optimal sets of predictors.
Psel
probability of selection of each variable as the proportion of times that is selected by AUC-RF method.
References
Calle ML, Urrea V, Boulesteix A-L, Malats N (2011) "AUC-RF: A new strategy for genomic
profiling with Random Forest". Human Heredity. (In press)
See Also
OptimalSet, AUCRF, randomForest.
Examples
# Next steps take some time
# load included AUCRF result example:
# data(fit)
# call AUCRFcv process:
# fitCV <- AUCRFcv(fit)
# The result of this example is included:
data(fitCV)
summary(fitCV)
plot(fitCV)
Results
R version 3.3.1 (2016-06-21) -- "Bug in Your Hair"
Copyright (C) 2016 The R Foundation for Statistical Computing
Platform: x86_64-pc-linux-gnu (64-bit)
R is free software and comes with ABSOLUTELY NO WARRANTY.
You are welcome to redistribute it under certain conditions.
Type 'license()' or 'licence()' for distribution details.
R is a collaborative project with many contributors.
Type 'contributors()' for more information and
'citation()' on how to cite R or R packages in publications.
Type 'demo()' for some demos, 'help()' for on-line help, or
'help.start()' for an HTML browser interface to help.
Type 'q()' to quit R.
> library(AUCRF)
Loading required package: randomForest
randomForest 4.6-12
Type rfNews() to see new features/changes/bug fixes.
AUCRF 1.1
> png(filename="/home/ddbj/snapshot/RGM3/R_CC/result/AUCRF/AUCRFcv.Rd_%03d_medium.png", width=480, height=480)
> ### Name: AUCRFcv
> ### Title: Repeated cross validation of the AUC-RF process.
> ### Aliases: AUCRFcv fitCV
>
> ### ** Examples
>
> # Next steps take some time
>
> # load included AUCRF result example:
> # data(fit)
>
> # call AUCRFcv process:
> # fitCV <- AUCRFcv(fit)
>
> # The result of this example is included:
>
> data(fitCV)
> summary(fitCV)
Number of selected variables: Kopt= 32
AUC of selected variables: OOB-AUCopt= 0.7787711
AUC from cross validation: 0.759109
Importance Measure: MDG
Selected.Variables Importance Prob.Select
1 SNP9 15.047305 1.00
2 SNP4 12.912120 1.00
3 SNP3 10.486599 1.00
4 SNP7 9.767075 1.00
5 SNP8 9.283819 1.00
6 SNP2 9.043039 1.00
7 SNP6 8.743129 1.00
8 SNP10 8.465736 1.00
9 SNP5 7.844703 1.00
10 SNP1 7.533021 1.00
11 SNP369 2.677609 0.35
12 SNP584 2.565316 0.19
13 SNP747 2.504847 0.09
14 SNP47 2.469360 0.26
15 SNP55 2.469196 0.14
16 SNP674 2.445041 0.24
17 SNP354 2.441501 0.04
18 SNP993 2.424503 0.16
19 SNP661 2.423057 0.51
20 SNP73 2.399690 0.03
21 SNP690 2.398267 0.56
22 SNP14 2.390978 0.05
23 SNP878 2.387848 0.50
24 SNP651 2.353301 0.00
25 SNP191 2.349521 0.36
26 SNP684 2.346010 0.16
27 SNP278 2.341461 0.06
28 SNP771 2.336632 0.04
29 SNP575 2.318485 0.71
30 SNP544 2.307716 0.61
31 SNP726 2.299561 0.13
32 SNP336 2.279044 0.07
> plot(fitCV)
>
>
>
>
>
> dev.off()
null device
1
>