subtable: Extract a subtable (or subset) from a table.

In hneth/i2ds: Introduction to Data Science

Extract a subtable (or subset) from a table.

Description

subtable yields a subtable of a table tbl by filtering or extracting a subset of table's dimensions and levels.

Usage

subtable(tbl, in_list = dimnames(tbl), out_list = NULL, quiet = FALSE)

Arguments

tbl	An original table to be reduced (as table).
in_list	A list specifying the subset of tbl to keep/include. Each list element is named after a dimension of tbl, and the value of that component is a vector of the names or a numeric index of the desired levels. Default: in_list = dimnames(tbl) (i.e., everything in tbl).
out_list	A list specifying the subset of tbl to drop/exclude. Each list element is named after a dimension of tbl, and the value of that component is a vector of the names or a numeric index of the desired levels. Default: out_list = NULL (i.e., nothing).
quiet	Boolean: Hide feedback messages? Default: quiet = FALSE (i.e., show messages).

Details

subtable provides a filter/slice function for tables, by specifying a positive subset (i.e., dimensions and levels to include) as in_list or a negative subset (i.e., dimensions and levels to exclude) as out_list (both as lists that specify dimension and level names, in tag and value format).

subtable assumes that dimnames(tbl) exist, but is flexible in allowing in_list and out_list to use both names or numeric indices for subsetting table levels.

As a key constraint, the subset of dimensions specified by in_list and out_list must still contain all dimensions (as excluding an entire dimension would leave no elements of a table) and yield a list with elements in the same order as dimnames(tbl) (to allow using do.call on what = "[", args to index/subset the array of tbl).

Value

A table.

Source

Based on the subtable function by Norman Matloff, The Art of R Programming (2011, pp. 131-134).

See Also

ctable for creating a contingency table; sublist for extracting subsets of a list; table and array for the underlying data structures.

Other array functions: add_dimnames(), ctable(), expand_freq_table(), flatten_array()

Examples

t <- datasets::Titanic

# Trivial case:
subtable(t, in_list = dimnames(t))  # same as t

# (a) Using level names:
subtable(t, in_list = list(Class = c("1st", "2nd", "3rd"),
                           Sex = "Female", Age = "Adult",
                           Survived = c("Yes")))

# Alternatives ways of obtaining same result:
subtable(Titanic, out_list = list(Class = c("Crew"),
                                  Sex = "Male", Age = "Child",
                                  Survived = c("No")))

subtable(Titanic, in_list = list(Class = c("1st", "2nd", "3rd"),
                                 Sex = "Female", Age = c("Adult", "Child"),
                                 Survived = c("Yes", "No")),
         out_list = list(Age = "Child", Survived = c("No")))

# (b) Using dim names and level numbers:
subtable(t, in_list = list(Class = 1:3, Sex = 2, Age = 2, Survived = 2))

# (c) Using only level numbers (note messages):
subtable(t, in_list = list(1:3, 2, 2, 2))

# (d) Using a mix of level names and numbers (note messages): 
subtable(t, in_list = list(1:3, "Female", 2, "Yes"))

# (e) Note: Different length of sub_dims than dimnamesI(tbl) yield ERRORs:
# subtable(t, in_list = list(1:3, "Female", 2, 2, 99))

# Any tbl dimensions (names/levels) not changed by in_list/out_list are included in full:
subtable(t, in_list = list(1:3, "Female", Survived = 2))  # both Age levels included
 

hneth/i2ds documentation built on Jan. 25, 2024, 2:22 p.m.