best.subsets function

F-test based best subset selection.

Adaptation of existing methods based on AIC/BIC.

best.subsets(model, nbest = 5, nvmax, digits, force.in = "NULL")

Arguments

• model: object class lm to select effects from.
• nbest: numeric indicating number of models to report of each size.
• nvmax: numeric maximum size of subsets to examine.
• digits: numeric giving number of digits in format of output.
• force.in: character vector indicating effects to keep in all models.

Details

F-based versions of built in subset method.

Returns

No return, only print.

Author(s)

Kristian Hovde Liland

Examples

data <- data.frame(y = rnorm(8),
                                   x = factor(c('a','a','a','a','b','b','b','b')),
                                   z = factor(c('a','a','b','b','a','a','b','b')))
mod <- lm(y ~ x + z, data=data)
best.subsets(mod)