Grouping/Aggregating by 2 Columns in R [duplicate]-CodePudding

I'm having a lot of trouble trying to figure out how to use the aggregate function in R and I'd appreciate some guidance. About the only way I could get aggregate to work was by following the syntax given here.

Given the data

data <- data.frame(alpha = c(1,1,1,1,2,2,2,2),
               beta = c(1,1,2,2,3,3,4,4),
               gamma = c(.1,.2,.3,.2,.4,.1,.3,.5))

I would like to aggregate by the columns 'alpha' and 'beta'. If I was just aggregating by 'alpha', for example, I could just do:

aggregate(x = data[ , colnames(data) != "alpha"], 
          by = list(data$alpha), 
          FUN = sum)

Trying

aggregate(x = data[ , colnames(data) != "alpha" | colnames(data) != "beta"], 
          by = list(data$alpha, data$beta), 
          FUN = sum)

didn't work. Could someone help me find the correct syntax?

Thanks!

CodePudding user response：

Using the formula interface might be easier here:

aggregate(gamma ~ alpha   beta, data = data, FUN = sum)

CodePudding user response：

how about:

library(dplyr)
data.frame(alpha = c(1,1,1,1,2,2,2,2),
           beta = c(1,1,2,2,3,3,4,4),
           gamma = c(.1,.2,.3,.2,.4,.1,.3,.5)) %>% 
  group_by(alpha) %>% 
  summarise(across(.fns = sum))

# A tibble: 2 x 3
  alpha  beta gamma
  <dbl> <dbl> <dbl>
1     1     6   0.8
2     2    14   1.3

CodePudding user response：

Using data.table

library(data.table)
setDT(data)[, .(gamma = sum(gamma)), .(alpha, beta)]