prepare_data: Prepares the data to be used by the 'mutual' function

In mutualinf: Computation and Decomposition of the Mutual Information Index

Prepares the data to be used by the mutual function

Description

Receives the data that is later used in the mutual function. Generates a data.table with the entry variables.

Usage

prepare_data(data, vars, fw = NULL, col.order = NULL)

Arguments

data	A tabular format object (data.frame, data.table, tibble). The data expected is microdata or frequency weight data for each combination of variables. The variables must be of "factor" class.
vars	A vector of variable names or vector of columns numbers contained in data. Also can be used "all_vars" to select all variables contained in data.
fw	Variable name or column number contained in data that contains frecuency weight for each combination of variables of the dataset. If this variable exists then the function will change its original name to fw. If this variable does not exist or is NULL, then the function will compute the frecuency weight given the combination of variables of vars and will create a new variable called fw. By default is NULL.
col.order	A variable name or vector of variables names contained in vars, or a column number or vector of column numbers contained in vars. Selects the columns to sort the dataset. By default is NULL.

Value

Returns a data.table of class "data.table" "data.frame" "mutual.data".

Examples

# Using some variable names in 'data' with explicit 'fw'.
my_data <- prepare_data(data = DF_Seg_Chile, vars = c("csep", "ethnicity", "school", "district"),
fw = "nobs")

# Using some column numbers in 'data' and explicit 'fw' as another column number.
my_data <- prepare_data(data = DF_Seg_Chile, vars = c(4, 5, 2, 3), fw = 11)

# Using all variables of 'data' with explicit 'fw'.
my_data <- prepare_data(data = DF_Seg_Chile, vars = "all_vars", fw = "nobs")

# Using some variable names in 'data' and 'fw' does not exist (in this case, the new 'fw' will
# be equal to 1 for all variable combinations as 'data' already has a frequency weights variable)
my_data <- prepare_data(data = DF_Seg_Chile, vars = c("csep", "ethnicity", "school", "district"))

# Using the 'col.order' option to sort data according to the 'csep' column.
my_data <- prepare_data(data = DF_Seg_Chile, vars = c("csep", "ethnicity", "school", "district"),
fw = "nobs", col.order = "csep")

# The class of the resulting object in all cases must be "data.table", "data.frame" and
# "mutual.data".
class(my_data)

mutualinf documentation built on June 22, 2024, 12:21 p.m.