Overwriting result with for loop in R

Question

I have a number of csv files and my goal is to find the number of complete cases for a file or set of files given by id argument. My function should return a data frame with column id specifying the file and column obs giving the number of complete cases for this id. However, my function overwrites the previous value of nobs in each loop and the resulting data frame gives me only its last value. Do you have any idea how to get the value of nobs for each value of id?

  myfunction<-function(id=1:20) {
  files<-list.files(pattern="*.csv")
  myfiles = do.call(rbind, lapply(files, function(x) read.csv(x,stringsAsFactors = FALSE)))  

   for (i in id) {  
   good<-complete.cases(myfiles)
   newframe<-myfiles[good,]
   cases<-newframe[newframe$ID %in% i,]
   nobs<-nrow(cases)  
  }
  clean<-data.frame(id,nobs)
  clean
 }

Thanks.

I don't know R, but you need to add nobs to a list or something similar, to be able to use that list as an argument to your data.frame function. — zoom

zx8754 zx8754 · Accepted Answer · 2016-02-06T17:04:30

We can do all inside lapply(), something like below (not tested):

myfunction <- function(id = 1:20) {
  files <- list.files(pattern = "*.csv")[id]

  do.call(rbind,
          lapply(files, function(x){
            df <- read.csv(x,stringsAsFactors = FALSE)
            df <- df[complete.cases(df), ]
            data.frame(ID=x,nobs=nrow(df))
            }
            )
          )  
}

Overwriting result with for loop in R

1 Answers