R/plot.condprob.R

In jhollist/condprob2: Package to conduct coniditional probability analysis

#' Function to plot condprob object
#'
#' This function plots a condprob object and creates a plot similar to Figure 1
#' in Hollister et al. (2008). This includes a simple scatter plot of the raw 
#' stressor vs the raw response, a cumulative distribution function for all of the 
#' data, the impaired sites, and the unimpaired sites, and a conditional probability 
#' plot with the conditional probability on the y-axis the stressor on the x-axis.  
#' 
#' @param cpobj An object of class condprob is required.  Data and parameters
#'                    from this object are used to create the plot
#' @param ... Used primarily to pass other ggplot2 plotting options to control title, 
#'            axis labels, colors, size, etc.
#' 
#' @references Hollister, J. W., J. F. Paul, and H. A. Walker (2008). CProb: 
#'             A Computational Tool for Conducting Conditional Probability 
#'             Analysis. Journal of Environmental Quality. 37(6):2392-2396.
#'             \href{http://dx.doi.org/10.2134/jeq2007.0536}{Link}
#' @export
#' @import ggplot2
#' @method plot condprob
plot.condprob <- function(cpobj, ...) {

  toplo <- data.frame(
    x = cpobj[[1]],
    y = cpobj[[2]],
    ylo = cpobj[[4]],
    yhi = cpobj[[5]]
  )
  xlb <- names(cpobj)[1]
  ylb <- names(cpobj)[2]
  
  p <- ggplot(toplo, aes(x = x, y = y)) + 
    geom_point() + 
    geom_line(aes(y = ylo), linetype = 'dashed') + 
    geom_line(aes(y = yhi), linetype = 'dashed') + 
    xlab(xlb) + 
    ylab(ylb)
  
  return(p)
  
}