Can we incrementally "add" data to existing series results?

[arclabch]

I'm looking for a way to update charts as new data become available, without recalculating everything each time. Is there a way to add new data to an already calculated series?

[DaveSkender]

Currently, no.

I have been thinking about this a bit for v2 in the roadmap (#218) and do see users occasionally trying to do one point at a time (#216). I think it makes sense in some use cases to do incremental calculations, especially for streaming models where you're dealing with real-time high frequency data. With that said, not all indicators can support it due to the way they are calculated, but I think many do.

In any case, it gets tricky because current period values are usually based on one or more previous values, so we'll have to have more in-memory storage of intermediate calculations for persistence and more memory transactions to make this work.

For now, the good news is that this library is quite fast, especially if you're using newer frameworks, and recalculating the whole time series is still feasible if you are not hitting memory or compute constraints.

[arclabch]

Thank you for the explanation. Appreciate it.

[DaveSkender]

Follow or continue this conversation in our discussion on streaming indicators for progress and information about pre-release package versions to support live and incremental quote sources.