Myujth

I currently have lots of excel files with the same fields but for different sectors. I'm trying to create a function that allows me to get the total mean from the "sulfate" field in all my excel sheets.

This is the code I currently have:

Mean_Pollution<-function(directory,pollutant,id = 1:332)
 directory <- c("001","002","003","004")
 for (x in directory)
 print(paste("Reading",x,"file"))
 temp = read.csv(paste(directory.path,x,".csv",sep = ""))
 print(paste("Finished reading",x,"file"))
 i = print(mean(temp$sulfate,na.rm = TRUE))
 

You seem to have everything you need and there are plenty of questions like this, but I'll provide an example. The assumption here is that everything is in 1 folder. I'll use my own folder:

setwd(dir = "C:/Users/Evan Friedland/Documents")
# Put some fake data in a fake folder
dir.create("Test_Folder")
setwd(dir = "C:/Users/Evan Friedland/Documents/Test_Folder")
n <- 5 # let's write 5 csv files
for(i in 1:n)
 write.csv(data.frame(madeupX = LETTERS[sample(1:24, 100, T)], sulfate = rnorm(100)), # fake data
 paste0(sprintf("%03d", i), ".csv")) # fake names

csvnames <- paste0(sprintf("%03d", 1:n), ".csv")

Now all you need to do to "stack" the mean of each file is initialize an empty vector, loop through them and save each result to an element of the vector.

means <- numeric(n) # intialize numeric type vector, n long
names(means) <- csvnames # name each element for fun
means <- sapply(1:n, function(x) # used sapply instead of a for loop but either is fine
 cat("+ ",csvnames[x],"n") # print which csv is running
 mean(read.csv(paste0(csvnames[x]))$sulfate) # return the mean of the sulfate col
 )
#+ 001.csv 
#+ 002.csv 
#+ 003.csv 
#+ 004.csv 
#+ 005.csv 
means # print results
#[1] 0.007859499 0.077447995 0.048796633 -0.101449790 0.224429258