QuantileCutoffs: Calculate Quantile Cutoffs

In bfatemi/easydata: Utility functions for easy data wrangling

Description

This function takes in a data.table, the column name of a measurement variable, the column name of a group variable, and a vector of quantiles in order to calculate quantile cutoffs for the measurement, by grouping.

Usage

QuantileCutoffs(DT, values, grouping, quants, bMelt = TRUE, round.digit = 3)

Arguments

DT	A data.table that contains the data
values	A character string that names the column of observations to calculate quantiles over
grouping	A character string that names the column containing the groups to calculate quantiles for
quants	A vector of quantiles defining cuttoffs (e.g. what is the value in values that defines the 30th percentile, by each group in the column group)
bMelt	A boolean indicating whether to output a long table (TRUE) or wide. Default behavior is to produce a long table, which is more suitable for plots
round.digit	An integer defining the decimal places to round the results to. Default behavior is 3

Details

The flags melt and round control rounding of the resulting quantiles and whether to return a long table (suitable for plotting) or a wide table (better for presentation of the table)

Value

A long or wide table showing quantile cutoffs for a given measurement by a group variable.

Examples

## - Example - 
## By US state, what is the 30th and 70th percentile for number of 
## jobs in that geographic area

library(data.table)

# Use data provided in this package
#  - change column class to numeric

DT <- copy(fiberCountyDem)
DT[, JobsCount := as.numeric(JobsCount)]

values <- "JobsCount"
grouping <- "State"
quants <- c(.3, .7)

# Set bMelt to false to receive a wide table for display purposes
#
QuantileCutoffs(DT, values, grouping, quants, bMelt=FALSE)

bfatemi/easydata documentation built on Oct. 7, 2019, 4:35 p.m.