QTSBE is a robust, open-source platform designed for backtesting quantitative trading strategies. At its core, it provides a powerful Python-based API built with Flask, offering extensive endpoints for seamless integration with existing trading systems and projects.
| Section | Field | Goal | Analysis | Reference |
|---|---|---|---|---|
| Positions | average_ratio | Average ratio of individual positions. | > | https://www.investopedia.com/terms/r/returnoninvestment.asp |
| Positions | final_cumulative_ratio | Total compounded return across all positions. | > | https://www.investopedia.com/terms/c/cagr.asp |
| Positions | average_position_duration | Mean duration (days) a position is held. | ~ | https://www.investopedia.com/terms/h/holdingperiod.asp |
| Positions | lowest_ratio | Lowest single‐trade ratio. | > | |
| Positions | lowest_ratio_buy_index | Index of the trade with lowest ratio. | info | |
| Positions | lowest_ratio_sell_index | Index when lowest ratio trade closed. | info | |
| Positions | biggest_ratio | Highest single‐trade ratio. | > | https://www.investopedia.com/terms/r/return.asp |
| Positions | biggest_ratio_buy_index | Index of trade with highest ratio. | info | |
| Positions | biggest_ratio_sell_index | Index when highest ratio trade closed. | info | |
| Positions | biggest_position_duration | Longest trade duration (days). | ~ | https://www.investopedia.com/terms/h/holdingperiod.asp |
| Positions | lowest_position_duration | Shortest trade duration (days). | ~ | https://www.investopedia.com/terms/h/holdingperiod.asp |
| Positions | max_cumulative_ratio | Peak cumulative return. | > | https://www.investopedia.com/terms/c/cagr.asp |
| Positions | min_cumulative_ratio | Lowest cumulative return encountered. | > | |
| Positions | average_cumulative_ratio | Mean of cumulative return curve. | > | https://www.investopedia.com/terms/c/cagr.asp |
| Positions | lowest_cr_ratio | Lowest cumulative ratio reset value. | > | |
| Drawdown | max_drawdown | Worst peak‐to‐trough decline. | < | https://www.investopedia.com/terms/m/maximum-drawdown-mdd.asp |
| Drawdown | average_drawdown | Mean drawdown over the backtest. | < | https://www.investopedia.com/terms/d/drawdown.asp |
| Drawdown | max_drawdown_period | Dates of peak→trough for max DD. | < | https://www.investopedia.com/terms/m/maximum-drawdown-mdd.asp |
| Advanced | sharpe_ratio | Risk-adjusted return (σ). | > | https://www.investopedia.com/terms/s/sharperatio.asp |
| Advanced | sortino_ratio | Downside risk-adjusted return. | > | https://www.investopedia.com/terms/s/sortinoratio.asp |
| Advanced | volatility | Standard deviation of trade returns. | < | https://www.investopedia.com/terms/v/volatility.asp |
| Advanced | annualized_return | CAGR of the strategy. | > | https://www.investopedia.com/terms/c/cagr.asp |
| Advanced | calmar_ratio | Annualized return ÷ max drawdown. | > | https://www.investopedia.com/terms/c/calmarratio.asp |
| Advanced | recovery_factor | Final return ÷ max drawdown. | > | https://www.investopedia.com/terms/r/recovery-rate.asp |
| Advanced | win_rate | % of winning trades. | > | |
| Advanced | loss_rate | % of losing trades. | < | |
| Advanced | profit_factor | Gross profit ÷ gross loss. | > | |
| Advanced | expectancy | Average return per trade. | > | |
| Advanced | trade_frequency_per_year | Number of closed trades per year. | info | |
| Advanced | exposure_pct | % of time capital is exposed. | info | |
| Advanced | skewness | Asymmetry of return distribution. | ≈0 | https://www.investopedia.com/terms/s/skewness.asp |
| Advanced | kurtosis | Tail heaviness of returns. | ≈3 | https://www.investopedia.com/terms/k/kurtosis.asp |
| Advanced | VaR_95 | 5% worst-case daily return. | > | https://www.investopedia.com/terms/v/var.asp |
| Advanced | CVaR_95 | Average of worst 5% returns. | > | https://www.investopedia.com/terms/c/conditional_value_at_risk.asp |
| Advanced | consecutive_wins | Longest winning streak. | > | |
| Advanced | consecutive_losses | Longest losing streak. | < | |
| Advanced | max_drawdown_period_days | Days between peak and trough of max drawdown. | < | https://www.investopedia.com/terms/m/maximum-drawdown-mdd.asp |
| Advanced | time_to_recovery_days | Days needed to recover after max drawdown. | < |
>: Higher value is better (maximize)<: Lower value is better (minimize)≈0: Value close to zero is ideal (neutral skewness)≈3: Value close to 3 is ideal (normal kurtosis)~: No optimal value, depends on strategy or contextinfo: Informational only, not directly used for scoring
- Clone the repository:
git clone https://github.com/simonpotel/QTSBE- Deploy with Docker:
docker-compose up-
Strategy Development
- Base template:
api/strategies/default.py - Example implementation:
api/strategies/rsi_example - Create your strategy file with required functions (
buy_signal,sell_signal,Indicators)
- Base template:
-
API Deployment
python api/api.py-
API Documentation
- API Documentation is available via Postman.
- Import the collection located in
docs/postman/QTSBE_API_Collection.json. - See Postman Documentation for detailed instructions.
-
Automated Data Collection
- Configure:
config/data_cron.json - Launch collector:
python tools/data/cron.py
- Configure:
For more detailed information, please refer to the following guides:
- Strategy Development : How to build and add your own trading strategies.
- Configuration : Detailed explanation of all configuration files.
- Data Management : Understanding HDF5 storage and data structures.
- Tooling : Overview of data fetching and maintenance tools.
- Postman Guide : How to import and use the API collection.
QTSBE uses Pytest for automated integration testing.
| Endpoint | Test Objective |
|---|---|
/QTSBE/health |
Service health and external connectivity |
/QTSBE/get_tokens |
Listing of available HDF5 datasets |
/QTSBE/get_tokens_stats |
Live performance and metadata validation |
/QTSBE/get_strategies |
Discovery of local Python strategy modules |
/QTSBE/analyse |
Core backtesting engine execution |
/QTSBE/analyse_custom |
Dynamic execution of injected strategy code |
To run the tests simply execute:
pytestThis project is licensed under the MIT License. See LICENSE for details.
Warning
This software is provided for research and educational purposes only. Users should:
- Thoroughly review and test all code before implementation
- Be aware that documentation may not reflect the latest updates
- Understand that the developers assume no liability for financial losses or calculation errors
- Conduct their own risk assessment before using in live trading
For professional inquiries and support, connect with me on LinkedIn.