frank - specifying multiple columns from data.table in R

Question

I am quote confused about the frank function. The documentation here says:

Only for lists, data.frames and data.tables. The columns to calculate ranks based on. Do not quote column names. If ... is missing, all columns are considered by default. To sort by a column in descending order prefix a "-", e.g., frank(x, a, -b, c). The -b works when b is of type character as well.

so i have my data:

structure(list(product = c("Product 1", "Product 1", "Product 1", 
                           "Product 1", "Product 1", "Product 5", "Product 5", "Product 5", 
                           "Product 5", "Product 5"), policyID = c("A738-33", "A738-33", 
                                                                   "A738-33", "A738-33", "A738-33", "A738-33", "A738-33", 
                                                                   "A738-33", "A738-33", "A738-33"), startYear = c(2014, 
                                                                                                                               2015, 2016, 2017, 2018, 2014, 2015, 2016, 2017, 2018), total = c("30000", 
                                                                                                                                                                                                     "30000", "30000", "30000", "30000", "10000", "10000", "10000", 
                                                                                                                                                                                                     "10000", "10000"), daily = c("150", "150", "150", "150", "150", 
                                                                                                                                                                                                                                     "80", "80", "80", "80", "80")), class = c("data.table", "data.frame"
                                                                                                                                                                                                                                     ), row.names = c(NA, -10L), .internal.selfref = <pointer: 0x7feec50126e0>, sorted = "product")

I want to order this data by columns total and daily. So i have done this:

> setDT(testDT)
> frankv(testDT, totallimit, rbddaily, ties.method="dense")
Error in colnamesInt(x, cols, check_dups = TRUE) : 
  argument specifying columns specify non existing column(s): cols[1]='30000'

strangely enough, when i DO use quotations, exactly the opposite what the documentatoin says, I am getting results:

frankv(testDT, cols=c("totallimit", "rbddaily"), ties.method="dense")

I also tried to integrating thin into data.table, and another weird thing happened. From the 10 rows of data I had, i obtained a 100 rows.

testDT[,.(rank = frankv(testDT, cols=c("limit", "daily"), ties.method="dense")), by = c("policyID", "product", "startYear")]

What am i doing wrong and how can i fix this? The documentation is not of much help, maybe i am missing something...

the colnames used in your code are not matching the ones in your sample data? — Wimpel, Jun 23 '20 at 10:28

score 2 · Accepted Answer · answered Jun 23 '20 at 10:41

2

For frank you should not quote, but for frankv (the function you used) you should:

library(data.table)
frank(testDT, total, daily, ties.method="dense")

 [1] 2 2 2 2 2 1 1 1 1 1

frankv(testDT, cols=c("total", "daily"), ties.method="dense")

 [1] 2 2 2 2 2 1 1 1 1 1

answered Jun 23 '20 at 10:41

Ahorn

3,686
1
10
17

how come when i use this as `testDT[,.(rank = frank(testDT, totallimit, rbddaily, ties.method="dense")), by = c("policyID", "product", "startYear")]`suddenly i get 100 rows instead of 10? – Nneka Jun 23 '20 at 10:53
what do you want to achieve with this line of code? – Ahorn Jun 23 '20 at 10:57
i want to add a column to the `data.table` that indicates the rank of products by total and daily for every policy&startyear combination – Nneka Jun 23 '20 at 11:52
oh, i just realized how pointless my attempt was. This is what i wanted to achieve`testDT$rank = frank(testDT, totallimit, rbddaily, ties.method="dense")` – Nneka Jun 23 '20 at 12:25

frank - specifying multiple columns from data.table in R

1 Answers1