Description Usage Arguments Details Value Author(s) References Examples
The function globalboosttest
implements a permutation-based testing procedure to globally test the (additional) predictive value of a large set of predictors given that a small set of predictors is already available.
1 |
X |
A n x p matrix or data frame with observations in rows and variables in columns, whose additional predictive value has to be tested. |
Y |
Either a n-vector of type factor (if the prediction outcome is binary), or a numeric vector of length n (if the prediction outcome is numeric and uncensored), or a |
Z |
A n x q matrix or data frame with observations in rows and variables in columns, on which we want to condition. Note that q should be smaller than n. If |
nperm |
The number of permutations used to derived the p-value. |
mstop |
A numeric vector giving the number(s) of boosting steps at which the p-value has to be calculated. |
mstopAIC |
If |
pvalueonly |
Should the function return only the permutation p-value or also the risk for all numbers of boosting steps and all permutations? |
plot |
If |
... |
Further arguments to be passed to the |
See Boulesteix and Hothorn (2009) for details on the methodology.
If mstopAIC=TRUE
, the number of boosting steps is chosen from 1 to max(mstop)
independently of the specific values
included in the vector mstop
.
A list with the following arguments
riskreal |
A numeric vector of length |
riskperm |
A |
mstopAIC |
The number of boosting steps selected using the AIC-based procedure (if |
pvalue |
A numeric vector of length |
Anne-Laure Boulesteix (http://www.ibe.med.uni-muenchen.de/organisation/mitarbeiter/020_professuren/boulesteix/eng.html),
Torsten Hothorn (http://www.statistik.lmu.de/~hothorn/)
A. L. Boulesteix and Torsten Hothorn (2010). Testing the additional predictive value of high-dimensional data. BMC Bioinformatics 10:78.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 | # load globalboosttest library
library(globalboosttest)
# load the simulated data with binary outcome
data(simdatabin)
attach(simdatabin)
# Test with 25 permutations
test<-globalboosttest(X=X,Y=Y,Z=Z,nperm=25,mstop=c(100,500,1000))
# load the simulated data with survival outcome
data(simdatasurv)
attach(simdatasurv)
# Test with 25 permutations
test<-globalboosttest(X=X,Y=Surv(time,status),Z=NULL,nperm=25,mstop=c(100,500,1000),mstopAIC=FALSE)
|
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.