Error: Passing a dataframe through quantile function

Question

Having trouble. What I have is:

dec_amt <- function(x, k) format(round(x, k), nsmall=k)      # Formatting decimal places 

example.df <- data.frame(replicate(8,sample(0:100000,1000,rep=TRUE)))
names(example.df) <- c("AF", "CD", "CS", "ED", "LP", "PI", "RR", "TD")

probTab_test2 <- function(x = c(...), y=c(...), z=c(...))

{ 
  m.TABLE <- list()
  EXP <- list()
  PROB <- c(seq(.10, .90, .10), seq(.91,.99,.01), seq(.995, .999, .001))
  PERIOD <- dec_amt(1/(1-PROB), 2)

  for (i in 1:(length(z)))  { 

    if (length(z) == 1) 
    { 
      break 
    } 

    EXP <- quantile(example.df[,z[i]], PROB)
    EXP <- formatC(EXP, format='d', big.mark=',')
    m.TABLE <- list(data.frame(PERIOD, EXP))

    print(m.TABLE)
  }

  EXP <- quantile(example.df[,z], PROB)
  EXP <- formatC(EXP, format='d', big.mark=',')

  TABLE <- data.frame(PERIOD, EXP)

  return(TABLE)
}

probTab_test2(c("Consumer Products"), c("All Revenues"),c("TD", "LP"))

 Error in `[.data.frame`(x, order(x, na.last = na.last, decreasing = decreasing)) : 
  undefined columns selected

What I want is, if the length of the argument z is > 1 then for every element of z, I want it to create an 'EXP' column that I can bind into a dataframe (defined as m.TABLE) so at the end I would have a list of 'z' number of dataframes.

I feel like the quantile function is not happy about me passing through a dataframe instead of a vector, but not sure how to get around that in this loop. Suggestions would be great, happy to provide further information.

Note - feel free to disregard my x, y arguments - those will be used to call an outside function, but not noteworthy for this issue.

quantile() is not for dataframes. library("fortunes"); fortune(85) — jogo
– jogo, Commented Jun 18, 2017 at 17:52
Right, is there any way around this issue? Can I somehow extract each specified element of 'z' column from the example.df issue as a vector and pass it through quantile()? Right now, I'm just grabbing it as a dataframe. — S31
– S31, Commented Jun 18, 2017 at 17:55

LucyMLi · Accepted Answer · 2017-06-18 19:01:05Z

1

You can make use of the lapply function to loop over the variables listed in z and simplify the code:

probTab_test2 <- function(x = c(...), y=c(...), z=c(...)) { 
    PROB <- c(seq(.10, .90, .10), seq(.91,.99,.01), seq(.995, .999, .001))
    PERIOD <- dec_amt(1/(1-PROB), 2)
    m.TABLE <- lapply(as.data.frame(example.df[, z]),  function (vector) {
        quantiles <- quantile(vector, PROB)
        formatted.quantiles <- formatC(EXP, quantiles, format='d', big.mark=',')
        return(data.frame(PROB, formatted.quantiles))
  })
    return(m.TABLE)
}

For each variable listed in z, this function calculates the quantiles and creates a data frame for each variable listed in z. The as.data.frame is necessary so that lapply works even when length(z)==1.

answered Jun 18, 2017 at 19:01

LucyMLi

6574 silver badges14 bronze badges

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

Collectives™ on Stack Overflow

Error: Passing a dataframe through quantile function

1 Answer 1

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Your Answer

Sign up or log in

Post as a guest

Related