Perfectly Flat Cross Table by native xtabs

Description

This function makes the result of xtabs perfectly flat.

Usage

fxtabs(
  formula = ~.,
  data = parent.frame(),
  subset,
  sparse = FALSE,
  na.action,
  na.rm = FALSE,
  addNA = FALSE,
  exclude = if (!addNA) c(NA, NaN),
  drop.unused.levels = FALSE,
  remove_zeros = 0,
  print = 0,
  debug = 0
)

Arguments

Details

formula and data parameters function the same way as native xtab function.
In addtion, those parameters marked as "xtabs parameter", see stats::xtabs.

The core part of thus function is as follows:

 res <- ftable( xtabs( lhs~rhs,  data=data ) )

The above returns a matrix when lhs is a matrix. However, as shown in the examples, when lhs is a vector or missing, the result is a cross table.
Note that, by adjusting row.vars and col.vars options of ftable, it is possible to produce a completely flat table.

Value

a matrix of flat contingency table.

Examples

seed=1701
set.seed(seed)
n <- 50
x1 <- sample(1:3,n,replace=TRUE)
x2 <- sample(LETTERS[1:2],n,replace=TRUE)
x3 <- sample(letters[1:2],n,replace=TRUE)
c1 <- sample(c(0:3),n,replace=TRUE)
c2 <- sample(c(0:5),n,replace=TRUE)
data=data.frame(x1,x2,c1,c2)

# 3 by variables
fxtabs( cbind(c1,c2)~x1+x2+x3, data=data)
ftable(xtabs( cbind(c1,c2)~x1+x2+x3, data=data))
fxtabs( c1~x1+x2+x3, data=data)
ftable(xtabs( c1~x1+x2+x3, data=data))
fxtabs( ~x1+x2+x3, data=data)
ftable(xtabs( ~x1+x2+x3, data=data))

# 2 by variables
fxtabs( cbind(c1,c2)~x1+x2, data=data)
ftable(xtabs( cbind(c1,c2)~x1+x2, data=data))
fxtabs( c1~x1+x2, data=data)
ftable(xtabs( c1~x1+x2, data=data))
fxtabs( ~x1+x2, data=data)
ftable(xtabs( ~x1+x2, data=data))

# single by variable
fxtabs( cbind(c1,c2)~x1, data=data)
ftable(xtabs( cbind(c1,c2)~x1, data=data))
fxtabs( c1~x1, data=data)
ftable(xtabs( c1~x1, data=data))
fxtabs( ~x1, data=data)
ftable(xtabs( ~x1, data=data))