Building Rust Backtests with an Engine or Streaming Node

Summary

This guide explains two Rust workflows for running Nautilus Trader backtests. The lower-level BacktestEngine approach assembles a simulated venue, instruments, in-memory market data, and a strategy before running the engine. It allows direct configuration of account and order-management settings, book type, balances, and fees. An EMA crossover strategy illustrates registration and execution.

The higher-level BacktestNode workflow reads data from a Parquet catalog and streams it in configured chunks. The guide walks through writing instrument and quote data, defining venue and data settings, building the node, adding a strategy to its engine, and running it. These examples teach setup and API structure rather than trading performance: they report no results or validation evidence. Users must supply data-loading logic and choose appropriate simulation assumptions, and the guide refers elsewhere for details on fill models and matching behavior.

Key ideas

• BacktestEngine provides direct control over simulated venues, instruments, data, strategies, and execution.
• BacktestNode loads catalog data and streams it in configurable chunks.
• Both examples configure a simulated venue and run an EMA crossover strategy.
• The high-level workflow requires catalog persistence and streaming support.
• The guide covers software setup, not strategy performance or the validity of simulation assumptions.

Tags

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.