I want to use OHLC time series at different frequencies (1min, 5min, etc.) - but from the same timeframe (ex: 13:00 to 14:00) - and the corresponding indicator values at each frequency to train a model with Tensorflow/Keras/Pandas. Is it possible and how?
Hi @Site_de_Niche & welcome to the forum.
There are multiples ways to achieve what you want to do. Solutions might depends on your original dataset/format/feed.
Slicing/manipulating data in Pandas might be the easiest way. You can find hints here
I agree with the previous poster, there are multiple ways to accomplish but one good way is to confirm your dataset has data for the smallest frequency you are trying to model, for instance, if a timeframe is (ex: 13:00 to 14:00) and the smallest frequency you are trying to model is 1 minute, that means your dataset should hold at the very least 60 seconds worth of data. Secondly, I would suggest a multi-modal model approach for this since it is different frequencies, and it would be more manageable. So essentially, you can use a LSTM model but in multi-modal model approach. You may want to include multi-outputs and multi-steps as well, inside each model. I hope this helps.
1 Like