CODEWITHSUNDEEP
r
adrCoder
adrCoder

Reputation: 3275

Data timestamped to minutes, keep data every 5, 30 minutes best way to implement it

I have some data that are timestamped every minute that look like this:

  date time_greece gmt_offset_greece price_greece time_and_date 
  gmt_offset_greece_test time_and_date_correct time_and_date_difference ID
1 2009-12-01  08:30:04.548 +2 2275.32 2009-12-01 08:30:04.548 2 2009-12-01 06:30:04 0 1

I want to perform different analyses for 5 minutes, 30 minutes intervals etc. At the moment I have created an ID based on the modulo operation between the row number and the 30 (I would do something similar for 5 minutes intervals etc.) 

statadata$ID <- seq.int(nrow(statadata))
statadata$ID <- seq.int(nrow(statadata)) %% 30

My question is, is there a more efficient way to implement this than the one I am currently using, that I haven't thought of / don't know?

Upvotes: 0

Views: 33

Answers (1)

MrGumble
MrGumble

Reputation: 5766

The package lubridate as a rounding function for date and datetimes, that can round to an arbitrary time unit, e.g. 5 minutes, 30 minutes, as well as floor and ceiling. With this you should be able to define your intervals as simple as lubridate::round(date_time_greece, '5 minutes'). As with all binning operations for data analysis, pay attention your groups. I.e. does your grouping/binning create many groups with just a single data point.

Upvotes: 0

Related Questions