Sharing Market Data Processing in Multi-Signal Strategies
Summary
This discussion addresses duplicated work in multi-signal trading systems when each signal independently processes ticks and bars using a bar generator and array manager. It recommends centralizing those components in the main strategy when signals use the same bar period, then passing updated bars and indicator data to each signal for its calculations.
It outlines three implementation patterns: signals consuming shared data directly, an event mechanism notifying signal modules after each bar update, and a composite strategy that aggregates outputs from separate signal strategies. The examples illustrate how RSI, CCI, and moving-average signals could be combined, but they provide no performance measurements or trading results. The advice is most applicable when signals share data frequency and inputs; differing periods or data requirements may still need separate processing. The discussion focuses on code structure and maintainability rather than signal validity or trading risk.
Key ideas
- A shared bar generator and array manager can eliminate repeated data processing when signals use the same bar period.
- Signal modules can receive bars and indicator state from a central strategy instead of maintaining duplicate processors.
- An event-driven design can notify multiple signal modules after a centralized bar update.
- A composite strategy can keep signal logic separate and aggregate each component's position output.
- The examples describe architecture choices but do not provide benchmarks or evidence of trading performance.
Tags
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.