In randrescastaneda/joyn: Tool for Diagnosis of Tables Joins and Complementary Join Features

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>"
)

library(joyn)
library(data.table)

This vignette will give you a brief overview of how you can use some auxiliary functions that joyn makes available to the user.

Verifying if dt is uniquely identified

One of the advantages of joyn is that you can perform one-to-one (1:1), one-to-many (1:m), many-to-one (m:1), and many-to-many (m:m) joins. is_id() is a function that might come in handy when you want to check whether your data table is uniquely identified by the variables you want to merge by. In fact this is what is_id() checks by default, returning either TRUE or FALSE depending on whether the data table is uniquely identified or not. Alternatively, you can set return_report = FALSE to get a summary of the duplicates.

x1 <- data.table(id = c(1L, 1L, 2L, 3L, NA_integer_),
                 t  = c(1L, 2L, 1L, 2L, NA_integer_),
                 x  = 11:15,
                 c  = c("a", "b", "a", "t", "d"),
                 c1 = c("h", "j", "k", "l", "y"))

y1 <- data.table(id = c(1,2, 4),
                 y  = c(11L, 15L, 16))

# Checking if x1 is uniquely identified by "id" with return_report = TRUE

is_id(dt = x1, 
      by = "id")

# Checking duplicates in x1 with return_report = FALSE

is_id(dt = x1, 
      by = "id", 
      return_report = FALSE)

Possible unique identifiers

In joyn, you can also search for variables which possibly uniquely identify your data table x using the possible_ids() function. For example,

# Identify possible unique identifier excluding variable t
possible_ids(dt      = x1, 
             exclude = "t")

# Identify possible unique identifier excluding character variables
possible_ids(dt      = x1, 
             exclude = "_character")

# Identify possible unique identifiers, excluding character variables but considering variable c1
possible_ids(dt      = x1, 
             exclude_classes = "character",
             include = "c1")

Verifying if data table is balanced

Additionally, joyn makes available to the user the is_balanced() function. This is instrumental in assessing the completeness of the data table within a specified group, i.e., if the table contains all the combinations of observations in the group. By default, is_balanced() will tell you if/if not the table is balanced. However, if you set return = "table", you will get a summary of the unbalanced observations. In other words, those combinations of elements between the specified variables that is not contained in the input table.

# Example with return = "logic", the default

is_balanced(df = x1,
            by = c("id", "t"))

# Example with return = "table"
is_balanced(df = x1,
            by = c("id", "t"), 
            return = "table")

Tabulating simple frequencies

Furthermore, joyn provides a function that generates simple frequency tables, so that you can easily have an overview of the distribution of values within your data tables.

# Tabulating frequencies of var `id`

freq_table(x     = x1, 
           byvar = "id")[]

# Removing NAs from the calculation

freq_table(x     = x1, 
           byvar = "id", 
           na.rm = TRUE)[]

randrescastaneda/joyn documentation built on Dec. 20, 2024, 6:51 a.m.