pollutantmean <- function(filedir, pollutant, csvid = 1:332) {
	pollutants = c()
	filenames = list.files(filedir)

	for(i in csvid) {
		filelocation = paste(filedir, "/", filenames[i], sep = "")
		filedata = read.csv(filelocation, header = TRUE)
		pollutants = c(pollutants, filedata[, pollutant])
	}
	pollutantsavg = mean(pollutants, na.rm = TRUE)
	pollutantsavg
}

complete <- function(filedir, csvid = 1:332) {
	csvidnums = c()
	nobs_vector = c()
	
	filenames = list.files(filedir)
	
	for(i in csvid) {
		filelocation = paste(filedir, "/", filenames[i], sep = "")
		filedata = read.csv(filelocation, header = TRUE)
		complete_cases = filedata[complete.cases(filedata), ]
		csvidnums = c(csvidnums, i)
		nobs_vector = c(nobs_vector, nrow(complete_cases))
	}
	data.frame(id = csvidnums, nobs = nobs_vector)
}

corr <- function(filedir, threshold = 0) {
	completes = complete(filedir, 1:332)
	completes_above_threshold = subset(completes, nobs > threshold)

	correl <- vector()

	filenames = list.files(filedir)

	for(i in completes_above_threshold$id) {
		filelocation = paste(filedir, "/", filenames[i], sep = "")
		filedata = read.csv(filelocation, header = TRUE)
		complete_cases = filedata[complete.cases(filedata), ]
		count = nrow(complete_cases)

		if(count >= threshold) {
			correl = c(correl, cor(complete_cases$nitrate, complete_cases$sulfate))
		}
	}
	correl
}