summary.pim: Plot very basic G.O.F. for a pim

In pimold: Probabilistic Index Models

Description

Plot very basic Goodness Of Fit for a pim

Usage

## S3 method for class 'pim'
summary(object,...)

## S3 method for class 'summary.pim'
print(x,...)

## S3 method for class 'pim'
vcov(object,...)

design.matrix(object, ...)

## S3 method for class 'pim'
design.matrix(object,...)

responses(object, ...)

## S3 method for class 'pim'
responses(object,...)

designcols(object, ...)

## S3 method for class 'pim'
designcols(object,...)

outcomeformula(object, ...)

## S3 method for class 'pim'
outcomeformula(object,...)

Arguments

object	pim object.
x	summary.pim object.
...	Ignored currently.

Value

For summary.pim: an object of class summary.pim. It holds the original call in an item call, and has an additional item coefficients, that is a matrix holding columns:

Estimate	The coefficient estimate.
Std. Error	Their standard error.
Z value	The standardized value (Test statistic for true coefficient zero).
Pr(>|z|)	The p-value.

For print.summary.pim: invisibly returns x

For vcov.pim: a (co)variance matrix

For design.matrix: The design matrix (of pseudo-observations)

For responses: The true responses (pseudo-observations)

For designcols: The orginal column names of the design matrix. These typically also contain the part of the formula used to build them, so indicate their source.

For outcomeformula: The left hand side of the formula used to. create the design matrix (i.e. to generate the pseudo-outcomes).

Examples

set.seed(1)
	myiris<-iris
myiris$xord<-ordered(sample(letters[1:3], nrow(myiris), replace=TRUE))
myiris$out<-runif(nrow(myiris))

irisprt<-myiris[sample(nrow(myiris), 10),] #10 random rows from iris
pim1<-pim(out~Sepal.Length, data=irisprt, link="logit")
summary(pim1)
vcov(pim1)
design.matrix(pim1)
responses(pim1)

pimold documentation built on May 2, 2019, 5:50 p.m.