Converting intraday to daily

[dr1ver1]

Hi,

I'm running my own analysis on the data in PST. Can anyone point me at the function(s) that convert intraday data to daily?
I'm currently using my own logic, but would be good to be consistent with the implemented approach.

Kind regards,
David

[robcarver17]

PST?

I just use pandas.

[dr1ver1]

Sorry - PST = pysystemtrade :-)
I'm looking for the logic that decides which intraday value to use when converting to daily. For instance system.rawdata.raw_futures_roll has intraday data

[robcarver17]

You know you have built a community when other people come up with an acronum for your own project and don't consult you, or even tell you...

Generally I use the last intraday value in the day. This will normally be the closing price, which I mark with a special time stamp of 23:00.

I should probably use some standard intraday to daily functions to enforce consistency, but it's mostly just code littered with pandas x.resample('1B').last()

The example you have highlighted is an exception where I take an average downstream. This is done by rawdata.daily_annualised_roll (annroll.resample("1B").mean())

The logic here is that we have a rough and highly unstable estimate of the roll differential which moves around during the day, and we want to take an average to smooth this out. Ultimately I take a ewma of this thing anyway before calculating the carry forecast, so this only helps in the corner case when the final value of the differential is Nan, but there are valid intraday values before then.

[dr1ver1]

Thanks! By coincidence, I've also been averaging raw_futures_roll and generally taking the last of others - for similar reasons.