summary.DSA {DSA}R Documentation

Summarization of the results of a call to the DSA routine

Description

Provides a useful summary of a call to the DSA routine.

Usage

summary.DSA(object, ...)

Arguments

object an object of class 'DSA'.
... currently ignored.

Value

An object of class 'summary.DSA'. Objects of class 'summary.DSA' have the following attributes:

model.selected a formula representing the final model selected by the DSA routine.
family a description of the link function for model.selected: gaussian indicates the indentity function while binomial indicates the logit function.
n.obs number of observations in the data frame data passed in to the DSA routine.
n.nonna number of complete observations on which model.selected is based (no missing values).
best.average.CVrisk minimum average cross-validated risk among all average cross-validated risks associated with all model sizes, interaction orders and dimension reduction levels considered by the DSA routine.
coefficients the coefficients of model.selected fitted on n.nonna observations from the learning set.
io.and.size the 'best' model size and order of interaction selected by cross-validation
average.CVrisk.model.selected average cross-validated risk for model.selected.
n.id number of independent experimental units in the data frame data passed in to the DSA routine.
n.id.nonna number of complete independent experimental units on which model.selected is based (no missing values).
nselected.vars the number of candidate variables after dimension reduction based on a user-specified rank cut-off or a rank cut-off selected by cross-validation.
ncandidate.vars the number of candidate variables truly considered, i.e. the number of variables whose ranks are lower than or equal to the largest cut-off value in rank.cutoffs.
n.vars the number of candidate variables considered pre-dimension reduction.
rank.cutoffs the rank cut-offs considered.
best.rank.cutoff NULL if the dimension reduction level is user-specified or the 'best' dimension reduction level if the dimension reduction step is based on cross-validation.
formula the original base formula passed in to the DSA routine.
moves a logical vector indicating whether deletion and substitution moves where permitted or inhibited in the model search.

Author(s)

James Bullard.

See Also

DSA.

Examples

library(DSA)

## an example using the state census data. (gaussian)  
data(state)
state.data <- as.data.frame(state.x77)
colnames(state.data) <- unlist(lapply(strsplit(colnames(state.data), " "),
                                      function(x) paste(x, collapse = "")))
res <- DSA(Murder ~ 1, data = state.data, maxsize = 5, maxsumofpow = 2,maxorderint = 2)
res
summary(res)

[Package DSA version 3.1 Index]