rbind inside nested for loops in R

Question

I need to create different dataframes, storing different dataframes with rbind inside a for loop, one for each i.

Here an example:

library(lubridate)

    DATA <- data.frame("ID"=c("01","01","02","02","03","03","03","04","04","04","05","05","05","06","06","06"),"x"=c("2009","2012","2013","2009","2012","2011","2013","2009","2010","2010","2011","2010","2009","2010","2011","2013"),"y"=c("a","a","a","b","c","a","a","c","b","b","c","a","c","a","b","c"))
    one <- c("a","b","c")
    two <- c("2009","2010","2011","2012")
listofdfs <- list()
    
for (i in one) {
  for (j in 1:(length(two)-1)) {
    B <- NULL
    A <- DATA[DATA$x %in% c(two[j],two[j+1]) & DATA$y==i ,]

    # keep the oldest date
      tmp<-as.Date(A$x,"%Y-%m-%d")
      
      if (length(tmp)>0)
      {
        A$tag <- 0
        names(tmp)<-1:nrow(A)
        id_x <- as.numeric(tapply(tmp[year(tmp)>two[j]],A$ID[year(tmp)>two[j]],function(x){
          id <- which.min(x)
          names(x)[id]
        }))
        A$tag[id_x] <- 1
      }
      
      # aggregate datasets over different years
      B <- rbind(B,A)
         }
   listofdfs[[i]] <- B
    }

It's almost want I want to obtain except for B that is not an aggregation of different A (A is overwritten each time).

I obtain a list like this:

$a
  ID    x y tag
2 01 2012 a   0
6 03 2011 a   0

$b
   ID    x y tag
15 06 2011 b   0

$c
   ID    x y tag
5  03 2012 c   0
11 05 2011 c   0

where 2010 is not present because of overwriting (which I don't want).

Any idea? Thanks!

Answer 1

Where you write

for (i in one) {
  for (j in 1:(length(two)-1)) {
    B <- NULL

should be

for (i in one) {
  B <- NULL
  for (j in 1:(length(two)-1)) {

with output

> listofdfs
$a
    ID    x y tag
1   01 2009 a   0
12  05 2010 a   0
14  06 2010 a   0
6   03 2011 a   0
121 05 2010 a   0
141 06 2010 a   0
2   01 2012 a   0
61  03 2011 a   0

$b
    ID    x y tag
4   02 2009 b   0
9   04 2010 b   0
10  04 2010 b   0
91  04 2010 b   0
101 04 2010 b   0
15  06 2011 b   0
151 06 2011 b   0

$c
    ID    x y tag
8   04 2009 c   0
13  05 2009 c   0
11  05 2011 c   0
5   03 2012 c   0
111 05 2011 c   0