R/findThetas.R

In tharris0924/btld: Triangular and Bimodal Triangular Linked Distributions

#' Determine modes of btld
#'
#' Using kernel density estimation, calculate the local maxima i.e. the modes
#'
#' @param xs input data.
#' @return modes
#'
#' @importFrom graphics hist
#' @export
#' @examples
#' x<-runif(1000,0,1)
#' findThetas(x)
findThetas<-function(xs){
  freq<- hist(xs, include.lowest=TRUE, plot=FALSE, breaks = 10)
  density<-freq$density
  modeidxs<-which(diff(sign(diff(density)))==-2)+1
  mids<-freq$mids
  theta<-mids[modeidxs]
  return(theta)
}