summary.bestglm: summary of 'bestglm' object

In bestglm: Best Subset GLM and Regression Utilities

Description

An analysis of deviance and a likelihood-ratio test with p-value. The p-value is greatly exagerated due to selection.

Usage

## S3 method for class 'bestglm'
summary(object, SubsetsQ=FALSE, ...)

Arguments

object	Output from the bestglm function
SubsetsQ	List best subsets of each size
...	optional arguments

Value

No value. Output to terminal only.

Author(s)

A.I. McLeod and C. Xu

See Also

bestglm, print.bestglm

Examples

data(znuclear)
summary(bestglm(znuclear))
#
#find statistical signficance of overall regression
data(Fires)
summary(bestglm(Fires, IC="BICq", t=1))

Example output

Loading required package: leaps
Fitting algorithm:  BIC-leaps
Best Model:
           df  deviance
Null Model 25  4.436699
Full Model 31 31.000000

	likelihood-ratio test - GLM

data:  H0: Null Model vs. H1: Best Fit BIC-leaps
X = 26.563, df = 6, p-value = 0.0001748

Note: in this special case with BICq with t =1only fitted model is returned.
With t=1, full model is fitted.
Fitting algorithm:  BICq(q = 1)-lm
Best Model:
            df  deviance
Null Model 456  842.0305
Full Model 516 1009.1015

	likelihood-ratio test - GLM

data:  H0: Null Model vs. H1: Best Fit BICq(q = 1)-lm
X = 167.07, df = 60, p-value = 4.897e-12

bestglm documentation built on March 26, 2020, 7:25 p.m.