👀 Views: 22 💬 Answers: 1 📅 Created: 2025-06-16

r data.table aggregation R

I'm prototyping a solution and I'm working on a project and hit a roadblock... I'm working with an scenario when trying to use `.SD` (Subset of Data) within a custom aggregation function inside a `data.table`. I want to compute a weighted average for groups, utilizing a custom function that takes into account both the values and their weights. However, the results seem incorrect, and I'm not sure why. Here’s what I’ve tried: ```R library(data.table) # Sample data dt <- data.table( group = c('A', 'A', 'B', 'B', 'C', 'C'), value = c(10, 20, 30, 40, 50, 60), weight = c(1, 2, 3, 4, 5, 6) ) # Custom weighted average function weighted_avg <- function(x, w) { sum(x * w) / sum(w) } # Using .SD in the aggregation dt[, .(weighted_avg = weighted_avg(value, weight)), by = group] ``` When I run the code above, I expect the output to show the correct weighted averages per group. However, I am noticing that the `weighted_avg` result for group 'A' returns `15` (which seems correct), for group 'B' it returns `37.5` (also correct), but for group 'C', it incorrectly outputs `55`. The expected value should actually be `57.5`. I've tried debugging the custom function by printing intermediate values within it, but the calculations seem to be fine. The version of `data.table` I'm using is `1.14.2`. Could there be something I'm overlooking with how `.SD` is being processed in the context of my function? Any insights would be greatly appreciated! How would you solve this?