How can i arrange columns as per given condition?

Question

My data set is :

S1   S2   S3    S4 
1    4    7     20
2    6    8     21
3    9    10    30

and i convert data as per condition that sort as per maximum difference value between each column value

data after the condition should be like this

S4    S2    S3    S1 
20    4     7     1
21    6     8     2
30    9     1     3

explanation: the difference between max and min value

for S4 is 30 - 20 =10

for S2 is 9-4 = 5

for S3 is 10-7 = 3

for S1 is 3-1 =2

and the columns are sorted as per the difference value of max and min.

Welcome to StackOverflow! Please read the info about [how to ask a good question](http://stackoverflow.com/help/how-to-ask) and how to give a [reproducible example](http://stackoverflow.com/questions/5963269). This will make it much easier for others to help you. — Sotos, Sep 06 '17 at 13:46

Sotos · Answer 1 · 2017-09-06T13:57:06.363

2

Here is an idea via base R,

df[order(sapply(df, function(i) max(i)-min(i)), decreasing = TRUE)]

which gives,

  S4 S2 S3 S1
1 20  4  7  1
2 21  6  8  2
3 30  9 10  3

If you want to change the column names to reflect the order you can use setNames, i.e.

setNames(df[order(sapply(df, function(i) max(i)-min(i)), decreasing = TRUE)], 
         paste0('column', seq_along(df)))

which gives,

   column1 column2 column3 column4
1      20       4       7       1
2      21       6       8       2
3      30       9      10       3

edited Sep 06 '17 at 13:57

answered Sep 06 '17 at 13:49

Sotos

51,121
6
32
66

and how can i give name to the columns same as per condition? – Rahul Semwal Sep 06 '17 at 13:51
What do you mean? – Sotos Sep 06 '17 at 13:52
like if i want to give name to second highest column from "S2" to "column2" and same for all columns i.e column1 etc.. – Rahul Semwal Sep 06 '17 at 13:54
Updated my answer – Sotos Sep 06 '17 at 13:57
and if i want names as column1, value_column2, 3_data_column, column_4_id then how can i do it? – Rahul Semwal Sep 06 '17 at 14:05
1

Thanks for your Help :D – Rahul Semwal Sep 06 '17 at 14:35
No problem. Please accept one of the answers (check tickmark next to it) so we can consider this question closed – Sotos Sep 06 '17 at 14:36

score 1 · Answer 2 · answered Sep 06 '17 at 14:07

1

Here is an option using tidyverse

library(tidyverse)
df1 %>% 
   summarise_all(funs(max(.)-min(.))) %>%
   unlist %>% 
   order(., decreasing = TRUE) %>% 
   select(df1, .)
#  S4 S2 S3 S1
#1 20  4  7  1
#2 21  6  8  2
#3 30  9 10  3

answered Sep 06 '17 at 14:07

akrun

874,273
37
540
662

and if i want names as column1, value_column2, 3_data_column, column_4_id then how can i do it? – Rahul Semwal Sep 06 '17 at 14:10
I get `Error in select(df, .) : unused argument (.)` – Sotos Sep 06 '17 at 14:12
@RahulSemwal You can use `rename_at` – akrun Sep 06 '17 at 14:14
@Sotos I used `dplyr_0.7.2` – akrun Sep 06 '17 at 14:14
Same here...`[1] ‘0.7.2’` – Sotos Sep 06 '17 at 14:16
@Sotos Not sure what would be the problem. I got it installed recently – akrun Sep 06 '17 at 14:17
hmm... Let me update the rest of the packages...Maybe something with dependencies... – Sotos Sep 06 '17 at 14:18
data is arranged but column name is still not there as per condition. – Rahul Semwal Sep 06 '17 at 14:24
@RahulSemwal IN your post `data after the condition should be like this` it shows with the same column names – akrun Sep 06 '17 at 14:26
No, i also want to change the column name as "column1", "value_column2", "3_data_column", "column_4_id". – Rahul Semwal Sep 06 '17 at 14:28
1

ahh...ok updated the rest of the packages and works great. Nice 1 :) – Sotos Sep 06 '17 at 14:31
1

Thanks i did it using setNames(df[order(sapply(df, function(i) max(i)-min(i)), decreasing = TRUE)], c("c1","c2","c3","c4")) – Rahul Semwal Sep 06 '17 at 14:34
1

Thank you for your Help, Nice :D – Rahul Semwal Sep 06 '17 at 14:35

score 0 · Answer 3 · answered Sep 06 '17 at 14:46

0

 df1[,order(sapply(df1,function(x)diff(range(x))),decreasing = T)]
   S4 S2 S3 S1
 1 20  4  7  1
 2 21  6  8  2
 3 30  9 10  3

answered Sep 06 '17 at 14:46

Onyambu

67,392
3
24
53

How can i arrange columns as per given condition?

3 Answers3