Creating new variables for multiple data frames in a for loop

Question

I have 8 data frames and I want to create a variable for each of this data frame. I use a for a loop and the code I have used is given below:

year <- 2001
dflist <- list(bhps01, bhps02, bhps03, bhps04, bhps05, bhps06, bhps07, bhps08)

for (df in dflist){
df[["year"]] <- as.character(year)
assign()
year <- year + 1
}

bhps01,...,bhps08 are the data frame objects and year is a character variable. bhps01 is the data frame for year 2001, bhps02 is the data frame for year 2002 and so on.

Each data corresponds to a year, so bhps01 corresponds to year 2001, bhps corresponds to 2002 and so on. So, I want to create a year variable for each one of these data. So, year variable would be "2001" for bhps01 data, "2002" for bhps02 and so on.

The code runs fine but it does not create the variable year for either of the data frames except the local variable df.

Can someone please explain the error in the above code? Or is there an alternative of doing the same thing?

Taufi · Accepted Answer · 2020-05-10 17:42:27Z

1

The syntax in the for loop is wrong. I am not entirely sure what you try to accomplish but let us try this

year = 2001 

A = data.frame(a = c(1, 1), b = c(2, 2))
B = data.frame(a = c(1, 1), b = c(2, 2))
L = list(A, B)

for (i in seq_along(L)) {
  L[[i]][, dim(L[[i]])[2] + 1] = as.character(rep(year,dim(L[[i]])[1]))  
  year = year + 1
}

with output

That is what you intend as output, correct?

In order to change the column name to "year" you can do

L = lapply(L, function(x) {colnames(x)[3] = "year"; x})

edited May 10, 2020 at 17:42

answered May 10, 2020 at 16:52

Taufi

1,5979 silver badges15 bronze badges

Sign up to request clarification or add additional context in comments.

7 Comments

user3571389 Over a year ago

I have tried this with the following code:

year = 2001 L <- list(bhps01, bhps02)  for (i in seq_along(L)) {   L[[i]][, dim(L[[i]])[1] + 1] = as.character(rep(year,dim(L[[i]])[1]))     year = year + 1 }

but it gives me the following error message: Error in [<-.data.frame(*tmp*, , dim(L[[i]])[1] + 1, value = c("2001", : new columns would leave holes after existing columns

user3571389 Over a year ago

I have 8 data frames namely, bhps01, bhps02, bhps03, bhps04, bhps05, bhps06, bhps07 and bhps08. Each data corresponds to a year, so bhps01 corresponds to 2001, bhps corresponds to 2002 and so on. So, I want to create a year variable for each one of these data. So, year variable would be "2001" for bhps01 data, "2002" for bhps02 and so on. I hope it is clear now?

Taufi Over a year ago

Sorry, there was a typo! Try this

year = 2001 L <- list(bhps01, bhps02) for (i in seq_along(L)) { L[[i]][, dim(L[[i]])[2] + 1] = as.character(rep(year,dim(L[[i]])[1])) year = year + 1 }

user3571389 Over a year ago

Hi the code runs this time, but still the variable "year" is not created for either of the data frame?

Taufi Over a year ago

No, the variable is probably V3. You can rename it, though, with the command L = lapply(L, function(x) {colnames(x)[3] = "year"; x}). Does it work now?

|

Omer F Yalcin · Accepted Answer · 2020-05-12 22:47:17Z

0

You take a copy of the dataframe from the list, and add the variable "year" to it, but then do not assign it anywhere, which is why it is discarded (i.e. not stored in a variable). Here's a fix:

year <- 2001
dflist <- list(bhps01, bhps02, bhps03, bhps04, bhps05, bhps06, bhps07, bhps08)

counter <- 0
for (df in dflist){
  counter <- counter + 1
  df[["year"]] <- as.character(year)
  dflist[[counter]] <- df
  year <- year + 1
}

If you want the original dataframes to be edited, you could assign the result back on the rather then into the list. This is a bit of an indirect route, and notice the change in creating the dflist with names. We create the df, and then assign it to the original name. For example:

year <- 2001
dflist <- list(bhps01 = bhps01, bhps02 = bhps02, bhps03 = bhps03, bhps04 = bhps04, bhps05 = bhps05, bhps06 = bhps06, bhps07 = bhps07, bhps08 = bhps08)

counter <- 0
for (df in dflist){
  counter <- counter + 1
  df[["year"]] <- as.character(year)
  dflist[[counter]] <- df
  assign(names(dflist)[counter], df)
  year <- year + 1
}

edited May 12, 2020 at 22:47

answered May 10, 2020 at 16:53

Omer F Yalcin

134 bronze badges

3 Comments

user3571389 Over a year ago

I have tried this code. The code runs fine, but the variable year is not created in the either of these dataframes.

Omer F Yalcin Over a year ago

That is because when you create a dataframe using bhps01, 02 etc., the changes are now in the list you created (dflist). If you want to have each of the individual dataframes to have the year variable, you might just assign them back. After the loop finishes: bhps01 <- dflist[[1]] ; bhps02 <- dflist[[2]] etc.

user3571389 Over a year ago

How can you automate bhps01 <- dflist[[1]] ; bhps02 <- dflist[[2]] etc. using loop or inside the same loop body?

Collectives™ on Stack Overflow

Creating new variables for multiple data frames in a for loop

2 Answers 2

7 Comments

3 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

7 Comments

3 Comments

Your Answer

Sign up or log in

Post as a guest

Related