CODEWITHSUNDEEP
rfunctiondataframecalculated-columns
Young Autobot
Young Autobot

Reputation: 81

Creating function to calculate between 2 columns in dataframe selecting largest number in column

I'm quite the novice when it comes to R and programming in general so any help would be very much appreciated. I have a large dataset with many rows with over 100 different IDs. Each 'ID' has 4 numbers. Each 'Number' has 5 records.

     ID Number     start       end   s.mean  Error
1   A2    1        61735  23342732   0.0314  2.04
2   A2    1     23345569  54962669  -0.0103  1.98
3   A2    1     54963958  55075062   0.4841  2.79
4   A2    1     55085141  65826284   0.0047  2.00
5   A2    1     65826928 115611498  -0.0241  1.96
6   A2    2        12784  17248573  -0.0037  1.99
7   A2    2     17248890  85480817  -0.0331  1.95
8   A2    2     85481399  89121495   0.0153  2.02
9   A2    2     89122081  89417610   0.3708  2.58
10  A2    2     89418929  89999062  -0.1826  1.76
11  A2    3    162626603 185477402  -0.0759  1.89
12  A2    3    185478957 189050664   0.0080  2.01
13  A2    3    189056732 192873807  -0.0985  1.86
14  A2    3    192874747 192882903   0.9053  3.74
15  A2    3    192886435 197896118  -0.0645  1.91
16  A3    1        61735  23342732   0.0314  2.04
17  A3    1     23345569  54962669  -0.0103  1.98
18  A3    1     54963958  55075062   0.4841  2.79
19  A3    1     55085141  65826284   0.0047  2.00
20  A3    1     65826928 115611498  -0.0241  1.96

I am wondering if it possible to create a function to subtract the 'start' from the 'end' of each record to determine which is the longest of each 'Number'. I was hoping for an output such as...

    ID   Number   Length 
1   A2    1       xxxxxx
2   A2    2       xxxxxx
3   A2    3       xxxxxx
4   A3    1       xxxxxx

Where 'xxxxxx' is the output of the longest length calculated.

Would it also be possible to select the largest 'Error' of each 'Number' using a function? Maybe having a similar output to the Length output above?

Not too sure how to tackle this. Again, any help would be much appreciated.

Upvotes: 2

Views: 37

Answers (1)

akrun
akrun

Reputation: 887891

We can use data.table. Convert the 'data.frame' to 'data.table' (setDT(df1)), grouped by 'ID', 'Number', get the difference of 'end', 'start' and find the max

library(data.table)
setDT(df1)[, .(Length = max(end - start)), .(ID, Number)]

Or with dplyr

library(dplyr)
df1 %>%
    group_by(ID, Number) %>%
    summarise(Length = max(end - start))

Upvotes: 1

Related Questions