I have loads of data (6500 files that I need to process and find yearly winter sums/averages/variance for. the format for my naming system is Syyyymmdd_rp. Each file is in the same file folder. I wanted to create a function that i could put a date and the statistical methods for calculations and I am having trouble starting. Keep in mind that the winter of 2014 spans back to September 2013. So essentially I want a function that I can put a start and end date into, as well as what method to calculate to rasters with (mean, sum, variance).


To select the files you could parse the numbers, but if the file names are as regular as you say (January is '01') I think you can also use their character representation ( '9' > '8' == TRUE )


selectedFilesFun <- function(files, start, end, fun) {
  b <- basename(files)
  b <- substr(b, 2, 9)
  i <- b > start & b < end
  f <- files[i]
  # if you have the files:
  # s <- stack(f)
  # calc(s, fun)

# get your files
# f <- list.files()
# here simulated: 

year <- rep(2000:2002, each=12)
month <- rep(c(paste0('0', 1:9), 10:12), 4)
f <- paste0('S', year, month, '15_rp')

x <- selectedFilesFun(f, '20010505', '20020202', fun=mean)

Not the answer you're looking for? Browse other questions tagged or ask your own question.