binned: Binned One-Way Data/Plot

In KO112/KO: Kevin O'Brien's Personal Package

Description

Calculate the (weighted) average value of a binned set of data, and plot it.

Usage

binned_one_way_data(
  x,
  yData,
  weight = rep(1, length(x)),
  scaleWeight = TRUE,
  type = "quantile",
  bins = 10
)

binned_one_way_plot(
  x,
  yData,
  weight = rep(1, length(x)),
  scaleWeight = TRUE,
  type = "quantile",
  bins = 10,
  fontSize = 10,
  showWeights = TRUE,
  plotly = FALSE,
  bgColor = "#CCDDFF",
  xlab = "Bins",
  ylab = "Response",
  wlab = "Weight",
  title = paste("One-Way Plot", if (!missing(xlab)) paste0("of ", xlab), if
    (!missing(ylab)) paste0("by ", ylab))
)

Arguments

x	The vector to perform binning by. If numeric, it will be binned according to type. If non-numeric, it will be binned according to all the unique values.
yData	A data.frame-like object (or numeric vector) whose numerical columns will be used to calculate average values by bin. Must have the same number of rows (or length) as the length of x.
weight	An (optional) vector of weights for weighted binning, and for taking weighted averages. Must be the same length as x.
scaleWeight	If TRUE, the weights will be scaled from 0 to 1 (i.e. as a percent).
type	The type of binning desired. Valid options are: r == raw, which will create a bin for each distinct value in x. e == equal, which will do the same as p == pretty. p == pretty, which will cut the range of x into equally sized bins, with the endpoints of each bin being chosen to be nice round numbers. q == quantile, which will cut x into bins according to weighted quantiles. m == min == minimum, which will try to ensure there are a minimum of bins observations in each bin.
bins	The number of bins to create (or, if type is in m == min == minimum, the desired minimum number of observations per bin).
fontSize	Size of the font to use in the plot.
showWeights	If TRUE, the weights plot will be shown as well.
plotly	Will return a plotly object instead of a ggplot one.
bgColor	Color to use for the background of the plot.
xlab	The x-axis label of the plot.
ylab	The y-axis label of the data section of the plot.
wlab	The y-axis label of the weight section of the plot.
title	The title of the plot.

Value

binned_one_way_data: a data.table holding the binned data.

binned_one_way_plot: a ggplot object, or a plotly object if plotly is TRUE.

Examples

diamonds <- data.table::data.table(ggplot2::diamonds)
binnedData1 <- binned_one_way_data(diamonds[, carat], diamonds[, .(x = x + 0.5, y, z)],
                                   diamonds[, price])
binnedData2 <- binned_one_way_data(diamonds[, carat], diamonds[, .(x = x + 0.5, y, z)],
                                   diamonds[, price], scaleWeight = FALSE)
binnedData3 <- binned_one_way_data(diamonds[, carat], diamonds[, .(x = x + 0.5, y, z)],
                                   diamonds[, price], type = "equal")
binnedData4 <- binned_one_way_data(diamonds[, carat], diamonds[, .(x = x + 0.5, y, z)], 
                                   diamonds[, price], bins = 5)
binnedPlot1 <- binned_one_way_plot(diamonds[, carat], diamonds[, .(x = x + 0.5, y, z)],
                                   diamonds[, price])
binnedPlot2 <- binned_one_way_plot(diamonds[, carat], diamonds[, .(x = x + 0.5, y, z)],
                                   diamonds[, price], scaleWeight = FALSE)
binnedPlot3 <- binned_one_way_plot(diamonds[, carat], diamonds[, .(x = x + 0.5, y, z)],
                                   diamonds[, price], type = "equal")
binnedPlot4 <- binned_one_way_plot(diamonds[, carat], diamonds[, .(x = x + 0.5, y, z)],
                                   diamonds[, price], bins = 5)

KO112/KO documentation built on Oct. 2, 2020, 9:21 a.m.