lm.cv: Get a fitted model selected by cross validation

In CSUV: Combined Selection and Uncertainty Visualiser (CSUV)

Description

Get a fitted model selected by cross validation

Usage

lm.cv(
  X,
  Y,
  intercept,
  fit.percent,
  num.repeat,
  method.names = NULL,
  num.core = 1,
  log.level = NULL
)

Arguments

X	covariates (n times p matrix, n: number of entries, p: number of covariates)
Y	response (vector with n entries)
intercept	TRUE to fit the data with an intercept, FALSE to fit the data without an intercept
fit.percent	percentage of observations used in fitting in cross validation. Each set of subsampling data will have (n times fit.percent) observations for fitting and n times (1-fit.percent) observations for calculating the mse
num.repeat	number of sets of subsampling data used in cross validation
method.names	vector of method names to be used in cross validation. Choose among "lasso", "elastic", "relaxo", "mcp" and "scad". Default is to use all methods listed above
num.core	number of cores to use. Default is 1 (i.e. no parallel running)
log.level	log level to set. Default is NULL, which means no change in log level. See the function CSUV::set.log.level for more details

Value

a list which includes the estimated coefficients (est.b) and the corresponding ordinary least square fit from stats::lm()

Examples

set.log.level(futile.logger::WARN)
X = matrix(rnorm(1000), nrow = 100)
Y = rowSums(X[,1:3])+rnorm(100)
cv.mod = lm.cv(X, Y, intercept = FALSE, fit.percent = 0.5, num.repeat = 50)
print(cv.mod$est.b)

CSUV documentation built on Oct. 23, 2020, 5:49 p.m.