You can use the following methods to calculate a cumulative sum by group in R:
Method 1: Use Base R
df$cum_sum
Method 2: Use dplyr
library(dplyr)
df %>% group_by(group_var) %>% mutate(cum_sum = cumsum(values_var))
Method 3: Use data.table
library(data.table)
setDT(df)[, cum_sum := cumsum(values_var), group_var]
The following examples show how to use each method in practice with the following data frame in R:
#create data frame df frame(store=rep(c('A', 'B', 'C'), each=4), sales=c(3, 4, 4, 2, 5, 8, 9, 7, 6, 8, 3, 2)) #view data frame df store sales 1 A 3 2 A 4 3 A 4 4 A 2 5 B 5 6 B 8 7 B 9 8 B 7 9 C 6 10 C 8 11 C 3 12 C 2
Example 1: Calculate Cumulative Sum by Group Using Base R
The following code shows how to use the ave() function from base R to calculate the cumulative sum of sales, grouped by store:
#add column to show cumulative sales by store
df$cum_sales #view updated data frame
df
store sales cum_sales
1 A 3 3
2 A 4 7
3 A 4 11
4 A 2 13
5 B 5 5
6 B 8 13
7 B 9 22
8 B 7 29
9 C 6 6
10 C 8 14
11 C 3 17
12 C 2 19
The new column called cum_sales displays the cumulative sum of sales, grouped by store.
Example 2: Calculate Cumulative Sum by Group Using dplyr
The following code shows how to use various functions from the dplyr package in R to calculate the cumulative sum of sales, grouped by store:
library(dplyr) #add column to show cumulative sales by store df %>% group_by(store) %>% mutate(cum_sales = cumsum(sales)) #view updated data frame df # A tibble: 12 x 3 # Groups: store [3] store sales cum_sales 1 A 3 3 2 A 4 7 3 A 4 11 4 A 2 13 5 B 5 5 6 B 8 13 7 B 9 22 8 B 7 29 9 C 6 6 10 C 8 14 11 C 3 17 12 C 2 19
The new column called cum_sales displays the cumulative sum of sales, grouped by store.
Example 3: Calculate Cumulative Sum by Group Using data.table
The following code shows how to use various functions from the data.table package in R to calculate the cumulative sum of sales, grouped by store:
library(data.table) #add column to show cumulative sales by store setDT(df)[, cum_sales := cumsum(sales), store] #view updated data frame df store sales cum_sales 1: A 3 3 2: A 4 7 3: A 4 11 4: A 2 13 5: B 5 5 6: B 8 13 7: B 9 22 8: B 7 29 9: C 6 6 10: C 8 14 11: C 3 17 12: C 2 19
The new column called cum_sales displays the cumulative sum of sales, grouped by store.
Note: All three methods produce the same result. However, the dplyr and data.table methods will tend to be quicker when working with extremely large data frames.
Additional Resources
The following tutorials explain how to perform other common calculations in R:
How to Calculate the Sum by Group in R
How to Calculate the Mean by Group in R
How to Calculate Standard Deviation by Group in R