Functions to assist in identifying and investigating outliers in water quality data.

Outlier.R

# function to calculate k of the max and min of a dataset
# k > 3 suggests data are 'far out'

Tukey_k <- function(x){
  my.quantile <- quantile(x, na.rm = TRUE)
  Q_25 <- my.quantile[2]
  Q_75 <- my.quantile[4]
  
  k_max <- as.vector((max(x, na.rm = TRUE) - Q_75)/(Q_75 - Q_25))
  k_min <- as.vector((Q_25 - min(x, na.rm = TRUE))/(Q_75 - Q_25))
  
  data.frame(k_max = k_max, k_min = k_min)
  
}

# Plot a histogram of the log of the data (with nice labels)
hist_log <- function(x,...){
  xname <-  deparse(substitute(x))
  hist(log10(x), breaks = 'sturges', xaxt = 'n', xlab = xname, main = '',..., col = 'blue')
  my_axTicks <- axTicks(1)
  my_ticks <- floor(min(my_axTicks)):ceiling(max(my_axTicks)) 
  axis(side = 1, at = my_ticks, labels = 10^my_ticks)
}

# QQ plot of logged data (with nice labels)
qq_log <- function(x){
  xname <- deparse(substitute(x))
  qqnorm(log10(x), yaxt = 'n', main = xname)
  my_axTicks <- axTicks(2)
  my_ticks <- floor(min(my_axTicks)):ceiling(max(my_axTicks)) 
  axis(side = 2, at = my_ticks, labels = 10^my_ticks, las = 2)
  qqline(log10(x), col = 'blue', lty = 2, lwd = 2)
}