Iterations with Loops and Functions

Question

I wanted to run a loop that read the value in a dataframe (data_rais), but I realised it might take days and I think is due to fact that I'm running a loop, and not a function. I tried several times to write a function that does the same as this loop, but I couldn't find a way to do so. My question is: Is it possible to transform this loop in a function? How?

   for(i in 1:nrow(data_rais)){
  if(is.na(data_rais$postal_code[i])){
    next()
  } else {
    data_rais$munic_name[i] = munics_code[row(munics_code)[which(munics_code$cods == data_rais$munic[i])], 1]
  }
}

munics_code looks like this:

  munics_code = tibble::tribble(
    ~municipio,~cods,
    'BELFORD ROXO', 261,
    'DUQUE DE CAXIAS', 250,
    'DUQUE DE CAXIAS', 251,
    'DUQUE DE CAXIAS', 252,
    'DUQUE DE CAXIAS', 253,
    'DUQUE DE CAXIAS', 254,
    'ITABORAÍ', 248,
    'ITAGUAÍ', 2380,
    'ITAGUAÍ', 2381,
    'ITAGUAÍ', 2382,
    'ITAGUAÍ', 2383,
    'ITAGUAÍ', 2384,
    'MAGÉ', 259,
    'MANGARATIBA',2386,
    'MANGARATIBA',2387,
    'MANGARATIBA',2388,
    'MARICÁ',249,
    'MESQUITA',2655)

And data_rais$postal_code is a column of a data_frame with numbers that may or may not start with the numbers in the cods column in munics_code. Something like...

data_rais = data.frame(postal_code = c(2049253, 2033069, 2293513, 2411920, 2284937, 2341811, 2008638, 
                                       2279827, NA, 2386135, 2441900, 2392889, 2332114, 2254610, 
                                       2114414, 2089509, 2351781, 2451466, 2111632, 2070417, 2079485, 
                                       2328146, 2200329, 2116103, NA, 2449114, 2231708, NA, 
                                       NA, 2194253),
                       munic_name = NA)

Note: I cannot delete the NAs, I don't want to lose them.

@Base_R_Best_R I wish I could, but it isn't possible since I'm dealing with sensitive data. — John P. S.
– John P. S., Commented Dec 24, 2019 at 14:28
If the data is sensitive, create a mock up data set with the same columns of interest. — Rui Barradas
– Rui Barradas, Commented Dec 24, 2019 at 14:33
"may or may not start with the numbers in the cods column in munics_code" does this mean you need partial matching of the numbers? — teofil
– teofil, Commented Dec 24, 2019 at 14:56

TobiO · Accepted Answer · 2019-12-24 14:47:27Z

3

I would suggest you use match

data_rais$munic_name = munic_code[[1]][match(data_rais$munic,munic_code$cods)]

to take care of entries when you already have a match in data_rais use the following:

data_rais$munic_name[!is.na(data_rais$postal_code)] = munic_code[[1]][match(data_rais$munic[!is.na(data_rais$postal_code)],munic_code$cods)]

Not sure if you need the second approach, but be careful with overriding original variables. If you're unsure add another variable and inspect the matching manually for a few entries.

edited Dec 24, 2019 at 14:47

answered Dec 24, 2019 at 14:33

TobiO

1,3811 gold badge9 silver badges25 bronze badges

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

2 Comments

TobiO Over a year ago

that's true :-) I rarely use this notation, and would normally suggest to use $ notation, also to be more consistent. I just got confused since the OP has different spellings of data frame and variable names... I updated my answer accordingly

TobiO Over a year ago

Sorry, I just realised that my original answer might override already placed entries, where you don't have a match in the data frame to match to but already have an entry in your original data_rais

R. Schifini · Accepted Answer · 2019-12-24 14:46:21Z

1

If I interpreted your code correctly, you are trying to set the data_rais$munic_name column to the corresponding municipio. This could be done with a merge:

df = merge(x = data_rais, y = munics_code, by.x = "postal_code", by.y = "cods", all.x = TRUE)

By doing a left merge (all.x = T) you'll preserve the NAs in data_rais. Assign the merge to data_rais if you want to add this column to it.

answered Dec 24, 2019 at 14:46

R. Schifini

9,3532 gold badges31 silver badges34 bronze badges

Collectives™ on Stack Overflow

Iterations with Loops and Functions

2 Answers 2

2 Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

2 Comments

Comments

Your Answer

Sign up or log in

Post as a guest