Replace Value in a Column Using a Loop and Custom Function - R - r

Question

I have a data.frame with a column (named "color") in which every value is "black." I also have created a function that can replace "black" with other colors depending on another column's value (the "growth" column value). I need to create a loop that uses this function to replace the values in the "color" column according to the "growth" value
# Create a function
check_it <- function(x)
if(x>500){
return("green")
} else if(x<0) {
return("red")
} else {
return("blue")
}
# Create a loop using check_it
for(x in 1:nrow(all_data)) {
...
# Given this hint:
# You can use 1:nrow(all_data) as a set of indices
# to do something like the following inside the loop:
# all_data[i, "color"] <-
# check_it( all_data[i, "growth"] )
Any suggestions?
SAMPLE DATA
| station_id | timestamp | growth.x | growth.y | color |
--------------------------------------------------------
| DB1 | 1/14/01 | 59.916 | 59.9164 | black |
--------------------------------------------------------
| DB1 | 1/14/02 | 316.128 | 316.128 | black |
--------------------------------------------------------
| DB1 | 1/14/03 | -12.456 | -12.456 | black |
--------------------------------------------------------
| DB1 | 1/14/04 | 537.443 | 537.443 | black |
--------------------------------------------------------
Thanks for the help! Thanks to the comments I was able to understand that my function wouldn't work without the proper arguments inserted (I just had "x") and didn't tell my function where to look for the "growth" value.
Here's the code I ended up using:
check_it <- function(x, )
if(all_data[x, "growth.x"] >500){
return("green")
} else if(all_data[x, "growth.x"] <0) {
return("red")
} else {
return("blue")
}
# Create a loop using check_it
for(x in 1:nrow(all_data)) {
all_data[x, "color"] <- check_it(x, all_data)
}

Answer 1

Well, of course there are plenty of solutions to your problem. But since you specifically requested a loop and provided your own function I tried to stick to what you've done so far as much as possible. You have however two growth-columns so I took the growth.y.
datf <- read.table(text="
station_id timestamp growth.x growth.y color
DB1 1/14/01 59.916 59.9164 black
DB1 1/14/02 316.128 316.128 black
DB1 1/14/03 12.456 12.456 black
DB1 1/14/04 537.443 537.443 black",
header = TRUE, stringsAsFactors = FALSE)
#I had to change your function a little:
check_it <- function(x, dat)
if(dat[x, "growth.y"] >500){
return("green")
} else if(dat[x, "growth.y"] < 0) {
return("red")
} else {
return("blue")
}
Now your loop-variable x corresponds to the row index of the data.frame and you're looping through it. Before that, this was not the case you just passed a number to your function.
#And finally the loop
for(x in 1:nrow(datf)){
datf[x, "color"] <- check_it(x, datf)
}
> datf
station_id timestamp growth.x growth.y color
1 DB1 1/14/01 59.916 59.9164 blue
2 DB1 1/14/02 316.128 316.1280 blue
3 DB1 1/14/03 12.456 12.4560 blue
4 DB1 1/14/04 537.443 537.4430 green
You should however consider to look at the *apply-function family.