Prepare stacked histogram data for plotting

Validates a patient-level or observation-level data frame and returns an hv_stacked object. Call plot.hv_stacked on the result to obtain a bare ggplot2 stacked (or proportional) histogram that you can decorate with colour scales, axis labels, and hv_theme.

Usage

hv_stacked(
  data,
  x_col = "year",
  group_col = "category",
  binwidth = 1,
  position = c("stack", "fill")
)

Arguments

data

A data frame.

x_col

Name of the numeric column to bin along the x-axis. Default "year".

group_col

Name of the column whose distinct values define the stacked groups. Will be coerced to a factor inside the aesthetic mapping. Default "category".

binwidth

Width of each histogram bin, in the same units as x_col. Default 1.

position

Bar position: "stack" (raw counts, the default) or "fill" (proportions that sum to 1 within each bin).

Value

An object of class c("hv_stacked", "hv_data"):

$data

The validated input data frame.

$meta

Named list: x_col, group_col, binwidth, position, n_obs, n_groups.

$tables

Empty list.

See also

plot.hv_stacked, sample_stacked_histogram_data

Examples

dta <- sample_stacked_histogram_data()

# 1. Build data object
sh <- hv_stacked(dta, x_col = "year", group_col = "category")
sh  # prints obs / group count
#> <hv_stacked>
#>   N obs       : 419  (3 groups)
#>   x / group   : year / category
#>   binwidth    : 1
#>   position    : stack

# 2. Bare plot -- undecorated ggplot returned by plot.hv_stacked
p <- plot(sh)

# 3. Decorate: fill/colour brewer palette, axis labels, theme
p +
  ggplot2::scale_fill_brewer(palette = "Set1", name = "Category") +
  ggplot2::scale_color_brewer(palette = "Set1", name = "Category") +
  ggplot2::labs(x = "Year", y = "Count") +
  hv_theme("poster")