QSTrader’s Planned Multi-Asset Backtesting and Portfolio Framework

Summary

This update describes a planned redesign of QSTrader from an equities-focused event-driven backtester into a system spanning research, simulation, paper trading, and live trading. Its proposed architecture separates alpha forecasts from portfolio construction: multiple alpha models generate forecasts, while a portfolio construction model combines them with risk and transaction cost assessments to produce target holdings and orders. The article names example approaches such as equal weighting, fixed-dollar allocation, and inverse-volatility weighting.

Other planned components include broker-managed master and sub-accounts with separate currencies, margin and borrowing costs, jurisdiction-specific commissions, cash transfers and dividends, and exchange calendars. A simulation timer is intended to decouple signal frequency from bar-data frequency. These features are presented as development plans and design goals, not as a performance study or complete implementation. The article gives no quantitative backtest results, and the proposed framework’s realism would depend on accurate market, cost, margin, and currency data.

Key ideas

• The proposed broker layer manages positions, sub-accounts, currencies, and consolidated account profit and loss.
• Alpha models produce asset forecasts, while portfolio construction combines forecasts with risk and transaction cost inputs.
• A simulation timer is designed to separate the frequency of strategy decisions from the frequency of available price bars.
• The planned system includes realistic margin, interest, commissions, cash flows, and dividends.
• The article describes intended capabilities and examples rather than validated trading performance.

Tags

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