selection: Selection Function for Genetic Algorithm for Variable...

In WaverlyWei/GA: Genetic Algorithm for model selection developed as a Final Project for Stat 243

Description

Selects independent variables to be bred by genetic algorithm based on AIC fitness function

Usage

selection(mm, model, parents, P)

Arguments

mm	Model matrix object with intercept and column for each independent variable specified in model
model	A formula object ( eg. data$y ~ x1 + x2^2 + x2:x3 ). Note: must specify data source for dependent variable
parents	Parents matrix of P rows indicating with 0 or 1 which variable selection
P	Population size for each generation

Examples

# simulate data
initData <- matrix( rnorm( 2500 , sd = c(1,  5, 7 , 100 , 40 ) ) , ncol = 5 , byrow = TRUE )
initOutcome <- 10 - 15 * initData[ , 1 ] + 2 * initData[ , 3 ] + 1.1 * initData[ , 5 ]
data <- data.frame( initData, initOutcome )

# define input parameters
P <- 30
parents <- initiation( C = 5 , P = P )
model <- data$initOutcome ~ X1 + X2 + X3 + X4 + X5
mm <- model.matrix( model , data = data )

# call selection function
selection( mm = mm , model = model , parents = parents , P = P )

WaverlyWei/GA documentation built on May 28, 2019, 6:41 a.m.