Filtering Down Data, then Adding Brackets - r

Question

Currently working with a data set about billionaires (small summary included below); I am looking to get them into three age brackets: "40 and under", "41 to 65", and "above 65", then what is the most common category (their kind of profession) of billionaire in each of the three age brackets.
I have tried to select the data down, and mutate or separate to create the brackets, but unsure what to do. The struck out lines are examples of things I have been trying.
load("bil.RData")
print(bil)
# A tibble: 2,614 x 22
age category citizenship company.name company.type country_code founded
<int> <chr> <chr> <chr> <chr> <chr> <int>
1 NA Financi… Saudi Arab… Rolaco Trad… new SAU 1968
2 34 Financi… United Sta… Fidelity In… new USA 1946
3 59 Non-Tra… Brazil Companhia B… new BRA 1948
4 61 New Sec… Germany Ratiopharm new DEU 1881
5 NA Financi… Hong Kong Swire new HKG 1816
6 NA Traded … Bahrain YBA Kanoo new BHR 1890
7 NA New Sec… Japan Otsuka Hold… new JPN 1921
8 NA Traded … Japan Sony new JPN 1946
9 66 Financi… Japan Mori Buildi… new JPN 1959
10 NA Traded … France Chanel new FRA 1909
# … with 2,604 more rows, and 15 more variables: `from emerging` <chr>,
# gdp <dbl>, gender <chr>, industry <chr>, inherited <chr>, name <chr>,
# rank <int>, region <chr>, relationship <chr>, sector <chr>, `was
# founder` <chr>, `was political` <chr>, wealth.type <chr>,
# worth_billions <dbl>, year <int>
bil %>%
select(age, category) %>%
arrange(age) %>%
filter(!is.na(age), !is.na(category)) %>%
group_by(age, category) %>%
#mutate(n = sum(age)) %>%
#separate(col = age, c("Under 40", "41-65", "Above 65")) %>%
print()
# A tibble: 2,158 x 2
# Groups: age, category [312]
age category
<int> <chr>
1 12 Financial
2 21 Financial
3 24 Financial
4 24 Financial
5 28 Non-Traded Sectors
6 28 Resource Related
7 29 Financial
8 29 Traded Sectors
9 29 New Sectors
10 29 New Sectors
Preferably, I am looking for a table with three rows (one per category, under 40, 41-65, 65+), and three columns (age_bracket, most common category, and n). Also, let me know the best way to include data sets on stackoverflow because this set is a bit large for dput() to be useful (I think).

Answer 1

I think there might be many ways to get what want.
Here are an example of using table1 :
library(tidyverse)
library(table1)
bil1<-bil %>%
mutate(age_group = cut(hp,
breaks = c(0, 40, 65, 110),
labels = c("< 40", "40 - 64", "65+")))
table1(~ category) | age_group, data = bil1)
You may also want to try many other packages such as arsenal and stargazer

Answer 2

To divide data into different buckets, use findInterval or cut and then use top_n to return top category in each bucket.
library(dplyr)
bil %>%
filter(!is.na(age)) %>%
group_by(group = findInterval(age, c(40, 65))) %>%
count(category) %>%
top_n(1, n)
This would just return 1, 2, 3 as labels, if you want to name the labels you can do
bil %>%
filter(!is.na(age)) %>%
group_by(group = c("40 and under", "41 to 65", "above 65")[
findInterval(age, c(40, 65)) + 1]) %>%
count(category) %>%
top_n(1, n)