get_ma: Obtain moving average map series

In leonawicz/mapmate: Make Still Image Sequences For Data Animations

Description

Obtain a moving average for monthly, annual or seasonal resolution data from monthly map data.

Usage

get_ma(x, type, season = NULL, size = 10, format = "table",
  use_mclapply = FALSE, mc.cores = 32)

Arguments

x	a data frame (or data table) or a list of these.
type	character, one of "monthly", "annual", or "seasonal".
season	NULL or character, one of "winter", "spring", "summer", or "autumn". Default is NULL.
size	number of years for the moving average window. Default 10.
format	return results as a single data frame with format="table" (default) or a as a list split either on unique year for seasonal and annual moving averages or on unique year and month combination for monthly moving averages.
use_mclapply	use mclapply from the parallel package (Linux). Defaults to FALSE.
mc.cores	number of CPUs for parallel processing when use_mclapply=TRUE.

Details

get_ma takes a single data frame or a list of data frames. A list is useful for example if the table is very large and is pre-split into a list of smaller data frames for parallel processing on a Linux cluster with many CPU cores available. If x is a data frame rather than a data table, it will be converted to a data table.

For each data frame get_ma computes and returns the moving or rolling average, after first summarizing monthly data to seasonal or annual averages if applicable. Winter begins in December and each season is three consecutive months (DJF, MAM, JJA, SON). Whether x is a data frame or list of data frames, the output can be returned as a single data frame or list of data frames

Parallel processing is Linux-only (uses mclapply) and the default of 32 CPU cores will have to be changed if you don't have that many. It is convenient for my usage and this package version is not intended for general use. Most other users will probably not be using parallel processing at all, in which case n.cores is ignored.

Value

returns a single data frame or a list of data frames.

Examples

data(monthlytemps)
res <- "seasonal" # annual, seasonal, monthly
season <- "winter" # winter, spring, summer, autumn
idx <- switch(season, winter=c(12,1,2), spring=3:5, summer=6:8, autumn=9:11)
if(res=="seasonal") monthlytemps <- dplyr::filter(monthlytemps, Month %in% idx)
get_ma(monthlytemps, res, season)
get_ma(list(monthlytemps, monthlytemps), res, season)
get_ma(monthlytemps, res, season, format="list")
## Not run: get_ma(list(monthlytemps, monthlytemps), res, season, use_mclapply=T, mc.cores=2)

leonawicz/mapmate documentation built on May 21, 2019, 5:09 a.m.