In TimothyKBook/rutilstb: A Package with Some Useful R Utilities

This is a package of various functions that I've made which I've found useful while working on different projects. They fall into two categories:

1. Functions I've found convenient while using Stata, and felt a need to port them into R.
2. Functions I've needed for a specific, narrow purpose.

library(rutilstb)

Functions Ported from Stata

The isid Function

The isid function can be used to detect whether or not a variable (or group of variables) has unique values. It has three uses:

Use #1: Vectors

a <- 1:5
b <- c(1, 1, 2, 3, 4)

isid(a)
isid(b)

Use #2: Data Frames

users <- data.frame(
  user_id = 1:5,
  first_name = c('Tim', 'Brandon', 'Brandon', 'Arya', 'Cersei'),
  last_name = c('Book', 'Stark', 'Tully', 'Stark', 'Lannister')
)

isid(users)

Use #3: Combinations of Variables in a Data Frame

isid(users, keys = c('first_name', 'last_name'))

The tabstat Function

The tabstat function can be used to view statistics based on categorical variables within a data frame.

tabstat('mpg', mtcars, by = 'cyl')
tabstat('mpg', mtcars, by = 'cyl', fns = list(median, IQR, length))

The mdesc Function

The mdesc function gives a description of missing values in a data frame, and where they are located.

df <- data.frame(
  a = 1:5,
  b = c(2, 3, NA, 4, 5),
  c = c(NA, 'x', 'y', 'z', NA)
)
mdesc(df)

Quick Functions of Narrow Use

The colMatcher Function

The colMatcher function tells the user if any two columns are identical.

df <- data.frame(
  x = 1:5,
  y = 101:105,
  z = 1:5,
  w = 101:105,
  v = letters[1:5]
)
colMatcher(df)
colMatcher(df, return_index = TRUE)

The dfSplit Function

The dfSplit function splits a data.frame into parts, either via a training/test split, or a K-fold split.

df <- data.frame(matrix(
  rnorm(100 * 4), 100, 4
))

split1 <- dfSplit(df, method = 'traintest', train_frac = 0.8)
lapply(split1, dim)

split2 <- dfSplit(df, method = 'traintest', train_n = 70)
lapply(split2, dim)

split3 <- dfSplit(df, method = 'kfold', K = 4)
lapply(split3, dim)

The fullRanker Function

The fullRanker function takes a matrix or data.frame and successively removes columns until it is of full rank.

X1 <- diag(4)
X2 <- cbind(X1, 1)

fullRanker(X1)
fullRanker(X2)

The unpack Function

The unpack function simply takes objects in a list, and unpacks them into the global (or otherwise unspecified) environment.

rm(list = ls())
ls()
mylist <- list(a = 1:5,
               b = letters[1:6],
               c = rnorm(7))
unpack(mylist)
ls()

TimothyKBook/rutilstb documentation built on May 9, 2019, 4:48 p.m.