ElasticNet: Wrapper function for glmnet

In ModelGood: Validation of risk prediction models

Description

Wrapper function for glmnet

Usage

ElasticNet(formula, data, nfolds = 10, ...)

Arguments

formula	Formula where the right hand side specifies the response and the left hand side the predictor matrix
data	A data frame in which formula is evaluated
nfolds	nfolds: number of cross-validation folds in cv.glmnet (default in function is 10)
...	passed on to glmnet

Details

This function first calls cv.glmnet and then evaluates glmnet at the hyper parameter which optimizes the cross-validation criterion.

Value

Object with class ElasticNet

Author(s)

Thomas A. Gerds <[e-mail]>

See Also

predictStatusProb

Examples

# Generate some data with binary response Y
 # depending on X1 and X2 and X1*X2
set.seed(40)
N <- 40
X1 <- rnorm(N)
X2 <- rbinom(N,1,.4)
X3 <- rnorm(N)
expit <- function(x) exp(x)/(1+exp(x))
lp <- expit(1 + X1 + X2 + X3)
Y <- factor(rbinom(N,1,lp))
dat <- data.frame(Y=Y,X1=X1,X2=X2,X3=X3)

efit <- ElasticNet(Y~X1+X2+X3,data=dat,family="binomial",alpha=0.1)
Brier(efit,verbose=FALSE)

Example output

Loading required package: glmnet
Loading required package: Matrix
Loading required package: foreach
Loaded glmnet 2.0-16

Sample size:  40 


Estimated Brier score in %
apparent 
   14.66 

Either newdata or apparent (learn data) performance.

ModelGood documentation built on May 2, 2019, 5 p.m.