nkaz001/hftbacktest
Hftbacktest
hftbacktest is a high-frequency trading backtesting framework and level 3 market data engine. It serves as a simulation environment for cryptocurrency trading bots and market-making strategies, utilizing a limit order book simulator to model precise market microstructures and liquidity.
The system differentiates itself through high-fidelity simulation components, including queue-position modeling to predict fill times and latency-aware execution to simulate network and exchange processing delays. It reconstructs order book states from level 2 and level 3 data and uses raw exchange trade and quote streams rather than aggregated candles.
The framework covers a broad range of capabilities including tick-based event simulation, limit order lifecycle management, and the analysis of market microstructure. It also includes tools for portfolio statistics tracking to calculate strategy metrics based on position states and pricing.
Features
- HFT Backtesters - Offers a high-fidelity simulation environment for testing high-frequency trading strategies using full tick-level data.
- Market Making Simulations - Provides a specialized environment for developing and validating automated liquidity provision strategies for cryptocurrency exchanges.
- Crypto Trading Bots - Functions as a specialized framework for cryptocurrency market making and arbitrage bots.
- Limit Order Lifecycle Managers - Handles the full lifecycle of buy and sell limit orders, including the enforcement of post-only restrictions.
- Limit Order Books - Implements a limit order book simulator that models the matching and queuing of buy and sell requests.
- Order Book Simulators - Models order queue positions and execution latencies to realistically simulate limit order fills.
- Queue Position Modeling - Tracks the relative position of limit orders within the order book to determine the exact moment of execution.
- Trading Simulations - Provides a controlled simulation environment for testing and executing high-frequency financial strategies.
- Tick Data Retrieval - Implements a system driven by the retrieval and processing of high-frequency, trade-by-trade tick data.
- Market Microstructure Analysis - Processes full order book snapshots and individual trade ticks to perform deep market microstructure analysis.
- Local Order Book Caches - Maintains a local representation of order book states by synchronizing Level-2 and Level-3 market data.
- L3 Market Data Engines - Analyzes individual order messages to simulate precise market microstructures and liquidity at the L3 level.
- Tick-Based Event Simulators - Processes market data and order book updates as a discrete sequence of events to simulate real-time trading.
- Backtesting Simulations - Uses an event-driven architecture to simulate high-frequency trading environments and validate strategies.
- Execution Latency Simulators - Provides high-fidelity simulation of network and exchange processing delays to accurately model execution latency.
- Portfolio Performance Trackers - Calculates performance metrics and strategy statistics by tracking current position states and mid-prices.
- Latency Impact Analysis - Evaluates the impact of network and processing delays on the overall profitability and execution of trading bots.
- Synchronization State Tracking - Implements state tracking to ensure strategy execution is synchronized with order confirmations for consistent portfolio states.
- Synchronous Order Executions - Implements synchronous order execution logic to ensure a strategy only proceeds after receiving an order confirmation.
- Trading and Backtesting - High-frequency trading backtester accounting for latency.
- Trading Frameworks - High-precision backtesting for HFT data using Numba.
Star history
How this analysis was created: This summary and feature list were written by an AI model that read the project's README and public documentation pages. Each feature links to the documentation it came from; stars, license and language come straight from the GitHub API. The model does not read the source code, and the analysis is refreshed when the project is re-analysed. Learn more on our About page.
