MXM (version 0.9.7)

BIC based forward selection with generalised linear models: Variable selection in generalised linear models with forward selection based on BIC

Description

Variable selection in generalised linear models with forward selection based on BIC

Usage

bic.glm.fsreg( target, dataset, wei = NULL, tol = 0, heavy = FALSE, 
robust = FALSE, ncores = 1)

Arguments

target

The class variable. It can be either a vector with binary data (binomial regression), counts (poisson regression). If none of these is identified, linear regression is used.

dataset

The dataset; provide either a data frame or a matrix (columns = variables, rows = samples). These can be continous and or categorical.

wei

A vector of weights to be used for weighted regression. The default value is NULL. It is not suggested when robust is set to TRUE.

tol

The difference bewtween two successive values of BIC. By default this is is set to 2. If for example, the BIC difference between two succesive models is less than 2, the process stops and the last variable, even though significant does not enter the model.

heavy

A boolean variable specifying whether heavy computations are required or not. If for exmaple the dataset contains tens of thousands of rows, it is advised to used memory efficient GLMs and hence set this to TRUE.

robust

A boolean variable which indicates whether (TRUE) or not (FALSE) to use a robust version of the statistical test if it is available. It takes more time than a non robust version but it is suggested in case of outliers. Default value is FALSE and is currently supported only by linear regression

ncores

How many cores to use. This plays an important role if you have tens of thousands of variables or really large sample sizes and tens of thousands of variables and a regression based test which requires numerical optimisation. In other cammmb it will not make a difference in the overall time (in fact it can be slower). The parallel computation is used in the first step of the algorithm, where univariate associations are examined, those take place in parallel. We have seen a reduction in time of 50% with 4 cores in comparison to 1 core. Note also, that the amount of reduction is not linear in the number of cores.

Value

The output of the algorithm is S3 object including: The output of the algorithm is S3 object including:

Details

Forward selection via the BIC is implemented. A variable which results in a reduction of BIC will be included, until the reduction is below a threshold set by the user (argument "tol").

See Also

fs.reg, lm.fsreg, bic.fsreg, CondIndTests, MMPC, SES

Examples

set.seed(123)

#simulate a dataset with continuous data
dataset <- matrix( runif(1000 * 50, 1, 100), ncol = 50 )

#define a simulated class variable 
target <- 3 * dataset[, 10] + 2 * dataset[, 20] + 3 * dataset[, 30] + rnorm(1000, 0, 5)
a1 <- bic.glm.fsreg(target, dataset, robust = FALSE, tol = 2, ncores = 1 ) 
a2 <- bic.glm.fsreg( round(target), dataset, robust = FALSE, tol = 2, ncores = 1 ) 

y <- target   ;   me <- median(target)  ;   y[ y < me ] <- 0   ;   y[ y >= me ] <- 1
a3 <- bic.glm.fsreg( y, dataset, robust = FALSE, tol = 2, ncores = 1 )