R/desc_num.R

In MathieuMarauri/explorer: Automatic data exploration with creation of markdown report.

#'
#' This function summarises a numeric vector. It returns table with descriptive statistics
#' and plots the density function, a boxplot and the cumulative density function of the
#' vector.
#'
#' @param x a numeric vector
#' @param plot logical, whether to plot the graph. Defaults to TRUE.
#'
#' @details Missing values are removed at the beginning and statistics are computed on the
#'   vector without missing data.
#'
#' @return a list with 3 tables. One with summary statistics (length, number of missing,
#'   min, max, mean, range, inter_quartile range, sd), another ne with the top 5 lowest
#'   and highest values and one with some percentiles (0, 0.05, 0.1, 0.25, 0.5, 0.75, 0.9,
#'   0.95, 1). If plot is true then a plot with the empirical density, a boxplot and the
#'   cumulative mepirical density is also returned.
#'
#' @export
#' @import ggplot2
#' @importFrom cowplot plot_grid
#'   
desc_num <- function(x, plot = TRUE) {
  # summary stats
  length <- length(x)
  x <- x[!is.na(x)]
  n_na <- length - length(x)
  percentiles <- quantile(
    x, 
    probs = c(0, 0.05, 0.1, 0.25, 0.5, 0.75, 0.9, 0.95, 1),
    na.rm = FALSE, 
    names = FALSE
  )
  range <- percentiles[9] - percentiles[1]
  range_iq <- percentiles[6] - percentiles[4]
  mean <- mean(x)
  sd <- sd(x)
  lowest <- sort(unique(x))[1:5]
  highest <- sort(unique(x), decreasing = TRUE)[1:5]
  # output
  summary_stat <- data.frame(
    Indicator = c('Length', 'Number of NAs', 'Min', 'Mean', 'Max', 'Range', 'Inter-quartile range', 
                  'Standard deviation'), 
    Value = c(length, n_na, percentiles[1], mean, percentiles[9], range, range_iq, sd)
  )
  extreme_values <- data.frame(
    value = c(paste(lowest, collapse = ', '), paste(highest, collapse = ', '))
  )
  rownames(extreme_values) <- c('Lowest values', 'Highest values')
  percentiles <- rbind(
    level = c('Min', 0.05, 0.1, 0.25, 'Median', 0.75, 0.9, 0.95, 'Max'),
    percentiles
  )
  rownames(percentiles) <- c('Level', 'Value')
  # plots
  # histogram
  hist <- ggplot() + 
    aes(x = x) + 
    geom_density(fill = 'grey30', alpha = .5) + 
    labs(y = 'Density') +
    theme(axis.text.x = element_blank(),
          axis.title.x = element_blank(),
          axis.ticks.x = element_blank())
  # boxplot
  boxplot <- ggplot() + 
    aes(x = 1, y = x) + 
    geom_boxplot() + 
    coord_flip() +
    theme(axis.text = element_blank(),
          axis.title = element_blank(),
          axis.ticks = element_blank())
  # ecdf
  plot_data <- data.frame(x = x, y = ecdf(x)(x))
  ecdf <- ggplot(data = plot_data, mapping = aes(x = x, y = y)) + 
    geom_line() + 
    labs(x = 'Data', 
         y = 'Cumulative density')
  
  graph <- cowplot::plot_grid(hist, boxplot, ecdf, 
                              align = "v", 
                              nrow = 3, 
                              rel_heights = c(1/2, 1/6, 1/3))
  if(plot) {
    print(graph)
  }
  invisible(list(summary_stat = summary_stat, 
                 extreme_values = extreme_values, 
                 percentiles = percentiles,
                 graph = graph))
}