This package contains methods for storing and manipulating collections of contingency tables, and for easily vectorizing functions which apply to a contingency table.
The basis of this is the class of object tables, which
contains a collection of numerical tables all of the same dimension.
Let’s create a collection of 10 contingency tables (in this case
probability tables), each of dimension 2x2x2.
library(contingency)## Loading required package: rjetab <- rprobMat(10, 2, 3)
tab## Group of 10 numeric tables of dimension 2x2x2
## First entry:
## , , 1
##
## [,1] [,2]
## [1,] 0.17618992 0.02182497
## [2,] 0.04537741 0.36982432
##
## , , 2
##
## [,1] [,2]
## [1,] 0.06995074 0.100333207
## [2,] 0.21492410 0.001575333The print method shows the first table in the list.
The tables are stored as a matrix as can be seen by using the
dim() function. Accessing particular rows of this matrix
return the appropriate tables:
tab[c(1,4,5),]## Group of 3 numeric tables of dimension 2x2x2
## First entry:
## , , 1
##
## [,1] [,2]
## [1,] 0.17618992 0.02182497
## [2,] 0.04537741 0.36982432
##
## , , 2
##
## [,1] [,2]
## [1,] 0.06995074 0.100333207
## [2,] 0.21492410 0.001575333However we can also specific elements of the tables using their co-ordinates, and (optionally) leaving the first entry blank:
tab[,1,1,]## Group of 10 numeric tables of dimension 2
## First entry:
## [1] 0.17618992 0.06995074The drop argument can be set to FALSE if
dimensions of length 1 should be retained:
tab[,1,1,,drop=FALSE]## Group of 10 numeric tables of dimension 1x1x2
## First entry:
## , , 1
##
## [,1]
## [1,] 0.1761899
##
## , , 2
##
## [,1]
## [1,] 0.06995074Some basic operations are predefined, such as taking the margin of each table, or calculating a conditional distribution.
margin(tab, 2:3) # margin of second and third dimensions## Group of 10 numeric tables of dimension 2x2
## First entry:
## [,1] [,2]
## [1,] 0.2215673 0.2848748
## [2,] 0.3916493 0.1019085conditional(tab, 2, 1) # second dimension conditional on first## Group of 10 numeric tables of dimension 2x2
## First entry:
## [,1] [,2]
## [1,] 0.6683178 0.4120643
## [2,] 0.3316822 0.5879357These can also be applied on an ordinary numerical array with the
expected effect. It can also be useful to calcuate conditional or other
functions but retain the placement of values in the same point as the
original table. For this purpose the functions margin2()
and conditional2() are available.
# as above but sequence of cells
margin2(tab, 2:3) # in table is retained## Group of 10 numeric tables of dimension 2x2x2
## First entry:
## , , 1
##
## [,1] [,2]
## [1,] 0.2215673 0.3916493
## [2,] 0.2215673 0.3916493
##
## , , 2
##
## [,1] [,2]
## [1,] 0.2848748 0.1019085
## [2,] 0.2848748 0.1019085conditional2(tab, 2, 1) ## Group of 10 numeric tables of dimension 2x2x2
## First entry:
## , , 1
##
## [,1] [,2]
## [1,] 0.6683178 0.3316822
## [2,] 0.4120643 0.5879357
##
## , , 2
##
## [,1] [,2]
## [1,] 0.6683178 0.3316822
## [2,] 0.4120643 0.5879357Some built-in functions are available. For example:
tab2 <- rprobMat(10,2,3)
kl(tab, tab2) # pairwise Kullback-Leibler divergence## [1] 1.0618544 0.3342060 0.9182598 1.2890683 0.8684732 0.3680129 0.8647746
## [8] 1.5426562 0.6994782 0.2843636 # mutual information between
mutualInf(tab, 2, 3) # second and third dimensions## [1] 0.0688704660 0.0005936809 0.0103282607 0.0211609209 0.0112977252
## [6] 0.0209483129 0.0120756909 0.0009720517 0.0135569948 0.0376785865mutualInf(tab, 2, 3, cond=1) # conditional mutual information## [1] 0.325450964 0.014315978 0.008692991 0.112810197 0.021386995 0.141698816
## [7] 0.059593111 0.034561512 0.020817647 0.085863122