Quant Strategy Backtester — DD < 2% RoI > 40% Sharpe 2.36

1. Overview

The complete backtest analysis for the strategy has been published using QuantStats, providing a deep dive into performance, risk, and return metrics. Click below to explore the full institutional-style interactive report (Sharpe, Sortino, Drawdowns, Rolling Returns, and much more): Click here to view the full QuantStats backtest report

---

2. Results Summary (QuantX V4.9 Full-Year 2024 Backtest)

Date Range: 2024-01-02 → 2024-12-31
Tickers: 32 (Dow 30 + additional large-cap stocks)

Metric	Value
Initial Capital	$1,000,000
Final Capital	$1,401,058
Total Return	+40.11%
Annualized Return	26.30%
Max Drawdown	–1.91%
Sharpe Ratio (annualized)	2.36
Trades Executed	3,308
Win Rate	66.63%
Avg Win / Loss	+$480.21 / –$345.07
Volatility (annualized)	11.13%
Profit Factor	1.72

Diagnostics Summary:

• Entries: 3,308
• Intraday exits: 55
• EOD closes: 3,253
• Z-score invalids (fails): 1,234,512
• Volatility filter fails: 27
• Trend filter fails: 568,989
• Position size fails: 48,938
• Cash availability fails: 353

---

3. Analytical Outputs

The following analytical components are provided as part of the QuantX V4.9 backtest package:

1. Equity Curve

   • Cumulative balance over time, starting from $1M base.
   • Equity initialization begins on 8 Jan 2024 after the first valid signal post warm-up period.

2. PnL Distribution

   • Histogram of per-trade profits/losses showing slight right-skew and light tails, consistent with mean-reversion payoff distribution.

3. Monthly Return Distribution

   • 11 out of 12 months were profitable; the worst drawdown was under 2%.

4. QuantStats Tear Sheet

   • Detailed return statistics, monthly returns, drawdown table, and benchmark-relative plots.
   • File: QuantX_V4.9_QuantStats_Report_Full.pdf.

5. Ticker Trade Charts

   • 32 individual PNGs showing price action with trade markers (▲ entries, ▼ exits).
   • Cleaned from zero-value anomalies.

6. Trade Log

   • QuantX_V4.9_Trades_Report.csv and .xlsx contain:
     • Time of Signal
     • Ticker
     • Entry/Exit prices
     • PnL
     • Z-score at signal
     • Position size

7. Equity Curve Data

   • QuantX_V4.9_EquityCurve.csv for custom analysis in Python or Excel.

---

4. Interpretation of Results

• The framework generated consistent low-drawdown performance, achieving ~26% annualized return with sub-2% drawdown.
• High trade frequency implies strong statistical validity and robustness of Z-score thresholds.
• Limited correlation with SPY confirms mean-reversion’s market-neutral nature.
• Drawdowns remain shallow, suggesting effective volatility-normalized position sizing and proper intraday stop management.

---

5. Future Scope and Expansion

1. Multi-Asset Extension:
   Expand universe to include ETFs, futures, or FX pairs with synchronized tick data for cross-asset mean reversion.

2. Machine Learning Enhancements:
   Integrate adaptive thresholding via regime detection (Hidden Markov Models or clustering-based volatility state identification).

3. Transaction Cost Modeling:
   Include realistic limit order fills using LOB simulation or dynamic spread modeling.

4. Execution Optimization:
   Test VWAP and TWAP-based execution for partial fill simulation.

5. Portfolio Optimization:
   Introduce dynamic capital allocation across tickers based on rolling Sharpe or drawdown-adjusted return.

6. Live Deployment:
   Integrate with broker APIs (e.g., Interactive Brokers) for paper/live trading environment.

---

6. Reproducibility

Environment

Tool	Version
Python	3.10
Pandas	2.1.1
NumPy	1.26
Matplotlib	3.8
QuantStats	0.0.60
pdfkit	1.0.0
wkhtmltopdf	0.12.6
Conda Environment	quantx_v49

Data

• Source: Proprietary minute-level OHLCV data for Dow 30 equities.
• Location: Data/1 Min Data/OHLC/
• Cached path (if using joblib): notebooks/cache_minute_data/joblib/__main__...

Reproduction Steps

1. Activate environment:
   conda activate quantx_v49

2. Run backtest notebook:
   notebooks/run_backtest.ipynb

3. Generate reports:
   notebooks/analysis.ipynb

4. Convert QuantStats HTML to PDF using pdfkit or browser print.

---

7. Repository Structure

Quant_Strategy_Backtester/
├── README.md
├── environment.yml
├── requirements.txt
├── data/
│ └── 1 Min Data/OHLC/
├── notebooks/
│ ├── run_backtest.ipynb
│ ├── analysis.ipynb
│ └── cache_minute_data/
├── src/
│ ├── strategy.py
│ ├── backtest.py
│ ├── execution.py
│ ├── reporting.py
│ └── utils.py
├── outputs/
│ ├── QuantX_Final_Backtest_Results.pdf
│ ├── QuantX_V4.9_QuantStats_Report_Full.pdf
│ ├── QuantX_V4.9_Trades_Report.xlsx
│ ├── QuantX_V4.9_EquityCurve.xlsx
│ ├── QuantX_V4.9_EquityCurve.png
│ ├── QuantX_V4.9_PnL_Hist.png
│ └── charts/
│ ├── chart_AAPL_all_trades.png
│ ├── chart_MSFT_all_trades.png
│ └── ...
└── docs/
└── backtest_performance_template.pdf

---

8. Conclusion

Strategy demonstrates the efficacy of a statistically disciplined, volatility-adjusted intraday mean-reversion framework.
The backtest confirms that a properly parameterized short-horizon strategy can achieve superior risk-adjusted returns while maintaining minimal drawdowns.

This repository serves as a complete and reproducible record of the strategy’s design, assumptions, implementation, and results, suitable for internal audit, investment committee review, or further research development.

---

9. License

Released under the MIT License.
Copyright © 2025 Alqama Ansari.

---

10. Citation

If referencing or using this codebase in research:

Ansari, A. (2025). QuantX V4.9: Intraday Mean-Reversion Strategy Backtest. GitHub Repository. https://github.com/ /Quant_Strategy_Backtester