R/dts1.R

In brisneve/ggplottimeseries: Visualisation of Decomposed Time Series with ggplot2

#' Dataframe for ggplottimeseries
#'
#' This function creates a data frame of decomposed time series.
#'
#' @param x a vector of the dates formatted as YYYY-MM-DD
#' @param y a vector of the observed time-series values
#' @param z frequency (the number of observations per unit of time)
#' @param type type of time series, either "additive" (default) or multiplicative
#'
#' @return This returns to a dataframe with the components of timeseries
#' \item{date}{date formatted as YYYY-MM-DD}
#' \item{observation}{observation of the timeseries}
#' \item{trend}{trend component of the timeseries}
#' \item{seasonal}{seasonal component of the timeseries}
#' \item{random}{random component of the timeseries}
#'
#' @examples
#' data(sst)
#'
#' x <- sst$date
#' y <- sst$sst
#' z <- 365.25 #number of days in a year
#' df <- dts1(x,y,z, type = "additive")
#'
#' head(df)
#'
#' @author Brisneve Edullantes
#'
#' @export



dts1 <- function(x,y,z, type = "additive"){

  if(!require(lubridate)){install.packages("lubridate"); library(lubridate)}
  if(!require(dplyr)){install.packages("dplyr"); library(dplyr)}

  #load ts data
  a <- ts(y,
          freq=z,
          start=decimal_date(ymd(as.Date.character(x[1]))))

  #decompose time series
  #collate date, data, trend, seasonal and random in a dataframe

  b <- decompose(a, type)
  c <- as.data.frame(cbind(observation = as.numeric(b$x),
                           trend = as.numeric(b$trend),
                           seasonal = as.numeric(b$seasonal),
                           random = as.numeric(b$random)))
  c$date <- x
  d <-dplyr::  select(c, date, observation, trend, seasonal, random)
}