Introduction

Financial markets are inherently volatile, requiring adaptive trading strategies to optimize returns while minimizing risks. StratFusion introduces an integrated approach by combining multiple technical indicator-based trading strategies into a single system. The project applies three stock trading strategies—RSI & Bollinger Bands, RSI & SMA Crossover, and Volume Spike Detection—to NVIDIA’s stock, leveraging high-frequency trading data at one-minute intervals. Additionally, an RSI-based options strategy was explored to expand the system’s adaptability.

Each strategy was individually optimized using grid search parameter tuning and backtested for performance. While these standalone approaches showed effectiveness under specific conditions, a unified system was developed to dynamically select and execute the most suitable strategy based on market conditions. The integrated model improves robustness, allowing it to be applied to diverse stocks and trading environments.

Methodology

Data Collection & Processing:

  • High-frequency stock data (1-minute intervals) for NVIDIA (Oct 29 – Nov 25, 2024) was sourced from Yahoo Finance.
  • Variables included price movements (Open, High, Low, Close), adjusted prices, and trading volume.
  • Preprocessing handled missing values and excluded non-trading hours.

Trading Strategies Implemented:

  1. RSI & Bollinger Bands Strategy – Identifies overbought (>70 RSI) and oversold (<30 RSI) conditions with Bollinger Bands for entry/exit signals.
  2. RSI & SMA Crossover Strategy – Uses short-term (50-period) and long-term (70-period) SMAs alongside RSI to identify trend shifts.
  3. Volume Spike Strategy – Detects large volume surges as signals for potential trend reversals or continuations.
  4. Unified System – Dynamically switches between strategies based on real-time market conditions for optimal performance.

Optimization & Backtesting:

  • Grid Search fine-tuned parameters such as RSI periods, Bollinger Bands multipliers, SMA lengths, and volume spike thresholds.
  • Performance evaluation metrics included total returns, Sharpe ratio (risk-adjusted return), and maximum drawdown (risk exposure).

Results & Analysis

Individual Strategy Performance:

  • RSI & Bollinger Bands Strategy: High returns in volatile markets but prone to false signals in low-volatility periods.
  • RSI & SMA Crossover Strategy: Strong trend-following ability but lagging signals in choppy markets.
  • Volume Spike Strategy: Effective in news-driven market surges but less reliable in normal conditions.

Unified System Performance:

  • Outperformed individual strategies by combining their strengths dynamically.
  • Reduced drawdowns and achieved superior risk-adjusted returns across market conditions.

Options Strategy Implementation

To expand StratFusion, options data from OPRA was analyzed using implied volatility (IV), Theta decay, and lognormal probability distributions. The strategy focused on optimizing out-of-the-money (OTM) options, where Theta decay is maximized, leading to better profitability.

Findings:

  • OTM options (strike prices: 134, 135, 136 for calls & 126, 127, 128 for puts) had the best profitability due to time decay advantages.
  • ITM options underperformed due to higher Delta sensitivity, leading to larger drawdowns and risk exposure.

Strategy Application to Microsoft Options:

  • Best performance on OTM puts (447.5, 450) with high profitability and strong risk control.
  • ATM and ITM options showed higher drawdowns, requiring tailored risk adjustments.

Conclusion

StratFusion successfully integrates multiple trading strategies into a robust, adaptable trading framework. The unified system outperforms individual strategies, demonstrating superior risk-adjusted returns. The options strategy proved most effective for OTM options, while ITM options require further refinements. The findings underscore the importance of dynamic strategy selection and parameter tuning in high-frequency trading environments.

This research highlights how integrating diverse trading methodologies enhances market adaptability, paving the way for more sophisticated algorithmic trading models.