Most common value (mode) by group in R

Question

I need to find the most common values for the unique names in column afor individual month. I know that the topic was already and I found a solution here: Is there a built-in function for finding the mode?, but I have a problem with multiple mode.

Solutions below:

Mode <- function(x) {
  ux <- unique(x)
  ux[which.max(tabulate(match(x, ux)))]
}

df<-data.frame(a=rep(c("a","b"),each=5),b=c(2,1,2,2,3,3,1,1,2,3), 
               c = c("Feb","Feb","Jan","Jan","Mar","Mar","Jan","Jan","Feb","Feb"))  
df %>% group_by(a,c) %>% summarise(d=Mode(b))

# A tibble: 6 x 3
# Groups:   a [2]
  a     c         d
  <fct> <fct> <dbl>
1 a     Feb       2
2 a     Jan       2
3 a     Mar       3
4 b     Feb       2
5 b     Jan       1
6 b     Mar       3

#When I want use:

Modes <- function(x) {
  ux <- unique(x)
  tab <- tabulate(match(x, ux))
  ux[tab == max(tab)]
}

df %>% group_by(a,c) %>% summarise(d=Modes(b))

#I get:
Error: Column `d` must be length 1 (a summary value), not 2

I expected:

1 a     Feb       2
2 a     Feb       1
3 a     Jan       2
4 a     Mar       3
5 b     Feb       2
6 b     Feb       3
7 b     Jan       1
8 b     Mar       3

If you could explain what you are doing and why, that would help. — user2974951, Feb 19 '20 at 11:53
Your `Modes` function returns more than one element if there are ties. Try `ux[tab == max(tab)][1]` as a (somewhat hackish) fix. — dario, Feb 19 '20 at 12:02
@dario It *intentionally* returns multiple values, the question already contains a simpler variant that returns a single value. — Konrad Rudolph, Feb 19 '20 at 12:03
Ok. I see. So do you want them returned as a list (`list(ux[tab == max(tab)])`) ? Or what do you want to do with these multiple values? — dario, Feb 19 '20 at 12:12

Aron Strandberg · Accepted Answer · 2020-02-19T12:47:06.427

3

You can do it like this:

library(dplyr)

df %>%
  count(a, b, c) %>%
  group_by(a, c) %>%
  filter(n == max(n)) %>%
  select(a, b, c)

Solution:

# A tibble: 8 x 3
# Groups:   a, c [6]
  a         b c    
  <fct> <dbl> <fct>
1 a         2 Feb  
2 a         1 Feb  
3 a         2 Jan  
4 a         3 Mar  
5 b         3 Mar  
6 b         1 Jan  
7 b         2 Feb  
8 b         3 Feb

edited Feb 19 '20 at 12:47

answered Feb 19 '20 at 12:27

Aron Strandberg

3,040
9
15

Most common value (mode) by group in R

1 Answers1