memisc: toDataFrame – R documentation

Become an expert in R — Interactive courses, Cheat Sheets, certificates and more!

toDataFrame

Convert an Array into a Data Frame

Description

to.data.frame converts an array into a data frame, in such a way that a chosen dimensional extent forms variables in the data frame. The elements of the array must be either atomic, data frames with matching variables, or coercable into such data frames.

Usage

to.data.frame(X,as.vars=1,name="Freq")

Arguments

`X`	an array.
`as.vars`	a numeric value or a character string. If it is a numeric value then it indicates the dimensional extend which defines the variables. If it is a character string then it is matched against the names of the dimenstional extents. This is applicable e.g. if `X` is a contingency table and the dimensional extents are named after the cross-classified factors. Takes effect only if `X` is an atomic array. If `as.vars` equals zero, a new variable is created that contains the values of the array, that is, `to.data.frame` acts on the array `X` like `as.data.frame(as.table(X))`
`name`	a character string; the name of the variable created if `X` is an atomic array and `as.vars` equals zero.

Value

A data frame.

Examples

berkeley <- Aggregate(Table(Admit,Freq)~.,data=UCBAdmissions)
berktest1 <- By(~Dept+Gender,
                glm(cbind(Admitted,Rejected)~1,family="binomial"),
                data=berkeley)
berktest2 <- By(~Dept,
                glm(cbind(Admitted,Rejected)~Gender,family="binomial"),
                data=berkeley)
Stest1 <- Lapply(berktest2,function(x)predict(x,,se.fit=TRUE)[c("fit","se.fit")])
Stest2 <- Sapply(berktest2,function(x)coef(summary(x)))
Stest2.1 <- Lapply(berktest1,function(x)predict(x,,se.fit=TRUE)[c("fit","se.fit")])
to.data.frame(Stest1)
to.data.frame(Stest2,as.vars=2)
to.data.frame(Stest2.1)
# Recasting a contingency table
to.data.frame(UCBAdmissions,as.vars="Admit")

memisc

Management of Survey Data and Presentation of Analysis Results

v0.99.27.3

GPL-2

Authors

Martin Elff (with contributions from Christopher N. Lawrence, Dave Atkins, Jason W. Morgan, Achim Zeileis)

Initial release

2020-11-18