AssessFitness: Quantify the 'fitness' of a given linear model

In stat243proj/GA: Appllies a modified genetic algorithm to some dataset, in order to assess what combination of predictor variables are best able to predict

Description

AssessFitness() determines 'fitness' of an invidivudal based on the quality of least squares fit of . The default for determining fitness is the Aikake Criteria Index but the user can supply their own custom-made fitness function that takes a linear model output object as its input.

Usage

AssessFitness(individual, response, predictors, user.family = "gaussian",
  userfunc = "AIC")

Arguments

individual	A binary vector of length C = # of potential co-variates in an individual. Co-variates to be used in the regression of response versus predictors are indicated by ones while the co-variates that are exlcuded are indicated by zeros.
response	The vector of response data (Y)
predictors	A data table or data frame of the co-variates (X)
user.family	Model family name to be passed to glm. Default is "gaussian"
userfunc	A fitness function that operates on a model that could be provided by the user. The default is the Aikake Information Criteria or "AIC".

Details

AssessFitness() returns a single fitness value.

Called from within Select()

Examples

\code{\link[GA]{FitnessFunction}}
\code{\link{glm}}
\code{\link{extractAIC}}

stat243proj/GA documentation built on May 15, 2019, 5:03 a.m.