superpc (version 1.12)

superpc.listfeatures: Return a list of the important predictors

Description

Return a list of the important predictor

Usage

superpc.listfeatures(data, 
                         train.obj, 
                         fit.red, 
                         fitred.cv=NULL,
                         num.features=NULL, 
                         component.number=1)

Arguments

data

Data object

train.obj

Object returned by superpc.train

fit.red

Object returned by superpc.predict.red, applied to training set

fitred.cv

(Optional) object returned by superpc.predict.red.cv

num.features

Number of features to list. Default is all features.

component.number

Number of principal component (1,2, or 3) used to determine feature importance scores

Value

Returns matrix of features and their importance scores, in order of decreasing absolute value of importance score. The importance score is the correlation of the reduced predictor and the full supervised PC predictor. It also lists the raw score- for survival data, this is the Cox score for that feature; for regression, it is the standardized regression coefficient. If fitred.cv is supplied, the function also reports the average rank of the gene in the cross-validation folds, and the proportion of times that the gene is chosen (at the given threshold) in the cross-validation folds.

References

• E. Bair and R. Tibshirani (2004). "Semi-supervised methods to predict patient survival from gene expression data." PLoS Biol, 2(4):e108.

• E. Bair, T. Hastie, D. Paul, and R. Tibshirani (2006). "Prediction by supervised principal components." J. Am. Stat. Assoc., 101(473):119-137.

Examples

# NOT RUN {
set.seed(332)

#generate some data
x <- matrix(rnorm(50*30), ncol=30)
y <- 10 + svd(x[1:50,])$v[,1] + .1*rnorm(30)
ytest <- 10 + svd(x[1:50,])$v[,1] + .1*rnorm(30)
censoring.status <- sample(c(rep(1,20), rep(0,10)))
censoring.status.test <- sample(c(rep(1,20), rep(0,10)))

featurenames <- paste("feature", as.character(1:50), sep="")
data <- list(x=x, 
             y=y, 
             censoring.status=censoring.status, 
             featurenames=featurenames)
data.test <- list(x=x, 
                  y=ytest, 
                  censoring.status=censoring.status.test, 
                  featurenames=featurenames)

a <- superpc.train(data, type="survival")
fit.red <- superpc.predict.red(a, 
                               data, 
                               data.test, 
                               .6)
superpc.listfeatures(data, 
                     a,  
                     fit.red, 
                     num.features=10)
# }