spareg (version 1.1.1)

screen_cor: Screening Coefficient Based on Correlation

Description

Creates an object class 'screencoef' using arguments passed by user, where the screening coefficient should be computed based on the correlation coefficient of response and each predictor separately.

Usage

screen_cor(..., control = list())

Value

object of class 'screencoef' which is a list with elements

• name (character)

• control (list of controls passed as an argument)

• generate_fun for generating the screening coefficient. This function should have arguments and y (vector of (standardized for Gaussian) responses), x (the matrix of standardized predictors) and a 'screencoef' object.

Arguments

...

includes arguments which can be passed as attributes to the 'screencoef' object

control

list of controls to be passed to the screening function

Details

Creates an object class 'screencoef' using arguments passed by user.

The function generate_fun relies on cor.

Arguments related to the screening procedure can be passed to the screen_cor() function through ..., and will be saved as attributes of the 'screencoef' object. The following attributes are relevant for spar and spar.cv:

• nscreen integer giving the number of variables to be retained after screening; if not specified, defaults to $2n$.

• split_data_prop, double between 0 and 1 which indicates the proportion of the data that should be used for computing the screening coefficient. The remaining data will be used for estimating the marginal models in the SPAR algorithm; if not specified, the whole data will be used for estimating the screening coefficient and the marginal models.

• type character - either "prob" (indicating that probabilistic screening should be employed) or "fixed" (indicating that a fixed set of nscreen variables should be employed across the ensemble); defaults to type = "prob".

• reuse_in_rp logical - indicates whether the screening coefficient should be reused at a later stage in the construction of the random projection. Defaults to FALSE.

Examples

example_data <- simulate_spareg_data(n = 200, p = 2000, ntest = 100)
spar_res <- spar(example_data$x, example_data$y, xval = example_data$xtest,
  yval = example_data$ytest, nummods=c(5, 10, 15, 20, 25, 30),
  screencoef = screen_cor(control = list(method = "kendall")))