@shravan-venkataraman

I am looking to backtest option strategies (i.e. long/put) using futures/index data, exactly the same question you mentioned at the beginning of this thread.

Are you able to use backtrader to backtest options based on Futures? If yes, have you find any similar article? Thanks!

@boblef In addition, I have already generated API token on my Oanda account and defined it and account id in my local code so you don't see them in the code I provided.

@backtrader said in Question on resampledata with 1min data to 60 mins:

The easiest to would be to substract 15 minutes to all your timestamps with a filter and then resample.

Playing around more with different resample mintues,

15 mins setup

cerebro.resampledata(data, timeframe=bt.TimeFrame.Minutes, compression=15)

result:

2019-02-26 09:30:00, 10794.00,10843.45,10789.55,10805.00,1605225 2019-02-26 09:45:00, 10805.30,10807.30,10775.05,10782.00,912975 2019-02-26 10:00:00, 10782.00,10786.25,10761.00,10764.80,825375 . . . 2019-02-26 15:00:00, 10878.05,10879.90,10817.00,10834.90,822975 2019-02-26 15:15:00, 10833.20,10844.00,10823.00,10829.00,563325 2019-02-26 15:30:00, 10828.05,10835.90,10821.85,10822.00,849300

30 mins setup

cerebro.resampledata(data, timeframe=bt.TimeFrame.Minutes, compression=30)

result:

2019-02-26 09:30:00, 10794.00,10843.45,10789.55,10805.00,1605225 2019-02-26 10:00:00, 10805.30,10807.30,10761.00,10764.80,1738350 2019-02-26 10:30:00, 10764.45,10777.35,10725.20,10775.20,2370375 . . . 2019-02-26 14:30:00, 10849.90,10866.00,10837.05,10864.75,931950 2019-02-26 15:00:00, 10865.80,10892.80,10817.00,10834.90,1656675 2019-02-26 15:30:00, 10833.20,10844.00,10821.85,10822.00,1412625

60 mins setup

cerebro.resampledata(data, timeframe=bt.TimeFrame.Minutes, compression=60)

result:

2019-02-26 10:00:00, 10794.00,10843.45,10761.00,10764.80,3343575 2019-02-26 11:00:00, 10764.45,10806.35,10725.20,10794.85,3613125 2019-02-26 12:00:00, 10793.00,10847.05,10792.75,10827.85,2194800 2019-02-26 13:00:00, 10828.00,10874.00,10815.75,10853.10,1884750 2019-02-26 14:00:00, 10851.15,10862.00,10827.00,10849.90,1200000 2019-02-26 15:00:00, 10849.90,10892.80,10817.00,10834.90,2588625 2019-02-26 15:30:00, 10833.20,10844.00,10821.85,10822.00,1412625

As per my findings, It's working fine for 5, 10, 15 mins!
It's not working for 30 mins and 60 mins.

When I says not working, first bar in 60 mins should be 10:15, instead of 10, and for 30 mins, first bar to be 9:45.

I'll dig into the code as well.

Thank you very much @backtrader. I didn't notice that thread. The solution works like a charm.

Oh thank you so much! Obvious mistake haha

@btr said in Programmatically extending a datafeed:

I needed a break and coffee. It was silly

Hilarious! Hope you had a nice break!

A sample of the returned value would be needed. Else use the native Quandl support available in the last release. It has only been tested with the WIKI dataset though.

See Docs - Data Feeds Reference

Glad it worked.

timeframe=1 seems anyhow odd because that is the actual value for bt.TimeFrame.Ticks and the data seems to be made up of 1-minute bars.

The platform will not complain (especially if no resampling is done), but you can see that later in a plot.

The suggestion would be to use: timeframe=bt.TimeFrame.Minutes (symbolic names seem better)