R/rts_data_example.R

In InstitutoInvestigacionesEconomicasPUCE/geortsBeta: Reconstruction of Geographical Time Series

#' @title Data Example for GeoRTS (Geo-reconstruction of Time Series)
#' @description This function generates a group of random data to use geoRts function.
#' @param m.ts is a interger that represents the number of time series "TS" to be calculated (i.e. \code{m.ts=dim(TS)[2])
#' @inheritParams geoRts
#' @return returns a list containing the original time series "TS0" (randomly generated), the positions and weights associated  with the original series (""positions.TS0" and "weights.TS0") and finally return the positions of the reconstruction of time series "positions.RTS0".
#' @export

rts_data_example = function(m.ts = 8, nx.rts = 5,ny.rts = 5){
	m=m.ts
	p= 50 # large of ts
	x = lapply(round(runif(m,3,20)), FUN = function(x){
		rnorm(p,mean = x,sd=1)
	})
	TS = as.data.frame(matrix(unlist(x) , nrow = p,ncol = m))
	TS = ts(data=TS, start = c(2000,1),frequency = 12)
	TS[15:20,] = NA

	positions.TS = data.frame(lon = rnorm(m,0,10),lat=rnorm(m,30,10))
	weights.TS = round(runif(m,1,10))
	#................................

	# positions.RTS = data.frame(lon = rnorm(n,0,10),lat=rnorm(n,30,10))
	# weights.RTS = round(runif(n,5,15))
	rangoX = abs( max(positions.TS$lon)-min(positions.TS$lon) )
	rangoY = abs( max(positions.TS$lat)-min(positions.TS$lat) )
	positions.RTS =expand.grid(lon =seq(from=min(positions.TS$lon)-0.05*rangoX,
										to=max(positions.TS$lon)+0.05*rangoX,
										length.out = nx.rts),
							   lat = seq(from=min(positions.TS$lat)-0.05*rangoX,
							   		  to=max(positions.TS$lat)+0.05*rangoY,
							   		  length.out = ny.rts))

	return(list("TS0" = TS,"positions.TS0" = positions.TS,
				"weights.TS0"=weights.TS,
				"positions.RTS0" = data.frame(positions.RTS)))
}