Home/ Questions/Q 8209697
In Process

The Archive Base Latest Questions

Editorial Team
  • 0
Asked: 2026-06-07T09:49:00+00:00

Hi I have a intra day data table with the following columns : date,

  • 0

Hi I have a intra day data table with the following columns : 

 date, stock_id, timestamp, price

First I added keys to order this properly : 

setkeyv( my_table, c('stock_id','date','timestamp'))

The data looks like :

     date      timestamp    stock_id      price       
 2011-01-04    1.294128e+12    7          3402.0     
 2011-01-04    1.294129e+12    7          3402.5     
 2011-01-04    1.294129e+12    7          3407.5

Now I would like to convert the stock_price to returns and log returns.

Could you please point to an efficient/elegant way to do this in R and data.table grouping without resorting to loops?

Many thanks in advance, I’m very new to R.

Share

Leave an answer

You must login to add an answer.

Need An Account,

1 Answer

  1. Editorial Team

    As it happens, you’ve asked something that is most easily done in the latest version (v1.8.1) on R-Forge repo using a new feature: := by group.

    How about (untested) :

    my_table[, logret:=c(NA,diff(log(stock_price))), by=stock_name]

    To install v1.8.1 from R-Forge without compiling from source, you need to be running R 2.15.0 or later and then just type :

    install.packages("data.table", repos="http://R-Forge.R-project.org")
    • 0