When working with time-series data, is it wrong to use different time-steps for the features and target?

Question

When working with time-series data, is it wrong to use daily prices as features and the price after 3 days as the target?

Or should I use the next-day price as a target, and, after training, predict 3 times, each time for one more day ahead (using the predicted value as a new feature)?

Will these 2 approaches give similar results?

Have you considered trying out both apporaches on your training data and comparing which one performs better? — don_pablito, Aug 23 '18 at 14:57
This question was also asked [here](https://datascience.stackexchange.com/q/36518). — nbro, Jan 01 '22 at 09:25

score 1 · Answer 1 · answered Aug 10 '18 at 00:32

I don't know what kind of price data you're dealing with. I suppose the order of the data matters a lot, so my suggestion would be:

Use LSTM as it handles time series better
You can predict 3 consecutive numbers from an RNN as the next three days' predictions
Try regression first, it is likely it will not work (or just flatten the curves, depend on your data noise), then classification is an easier approach
Don't forget normalization

When working with time-series data, is it wrong to use different time-steps for the features and target?

1 Answers1