efron2004: Diabetes Data from Efron et al. (2004)

In care: High-Dimensional Regression and CAR Score Variable Selection

Description

Diabetes data (10 variables, 442 measurements) as used in the study of Efron et al. (2004). The data is standardized such that the means of all variables are zero, and all variances are equal to one.

Usage

data(efron2004)

Format

efron2004$x is a 422 x 10 matrix containing the measurements of the explanatory variables (age, sex, body mass, etc.). The rows contain the samples and the columns the variables.

efron2004$y contains the response.

Source

The orginal data are available in the lars R package, see https://cran.r-project.org/package=lars. Note that this uses a slightly different standardization.

References

Efron, B., et al. 2004. Least angle regression (with discussion). Ann. Statist. 32:407–499. <DOI:10.1214/009053604000000067>

Examples

# load care library
library("care")

# load Efron et al. (2004) diabetes data set
data(efron2004)
dim(efron2004$x) # 442 10
colnames(efron2004$x)
length(efron2004$y) # 442

Example output

Loading required package: corpcor
[1] 442  10
 [1] "age" "sex" "bmi" "bp"  "s1"  "s2"  "s3"  "s4"  "s5"  "s6" 
[1] 442

care documentation built on Nov. 21, 2021, 5:06 p.m.