📈 Event-Driven Trading Strategy Using Mention Spikes & Price Momentum

Overview

This repository contains a quantitative trading framework designed to identify profitable trading signals based on mention spikes, price momentum, and now classical indicators like RSI, MACD, CVD, and Chaikin Oscillator. The framework supports hyperparameter tuning, visual signal analysis, and backtesting for event-driven trading strategies.

Data available here: 📂 Google Drive

---

🚀 Key Features

• 📊 Hyperparameter Optimization
• 📈 Multi-Horizon Event Study
• 📌 Statistical & Performance Metrics
• 🔎 Signal Filtering for Trading
• 🧠 Classical Technical Indicator Studies (RSI, MACD, CVD, Chaikin)
• 📊 Data Visualization & Insights
• ⏱️ Backtesting Interface to simulate and visualize event signals

---

🧠 New Additions

• 🔁 Backtesting Framework: Simulate event-based trades using run_study.py and symbol_data.py
• 🧮 Classical Trading Studies: Add custom indicator studies to augment or replace mention-driven events
• 🖼️ Visual Signal Explorer: Use plot_fcn.py to plot annotated trading signals

🔍 Example Signal Output

Example Strategy

Sharpe Ratio 2.75, Asset: Tesla, Entry Position: Long and Short Possible.

📂 Repository Structure

./data/                     # Datasets (QuiverQuant, Yahoo Finance, Mentions, etc.)
./event_study/              # Event study logic and hyperparameter search
    ├── event_studies.py
./preproc/                  # Data preprocessing
./utils/                    # Utility functions (plotting, stats)
./UpdateMentionData.py      # Script to update data sources

./backtesting/              # Backtesting framework
    ├── plotting/
        └── plot_fcn.py     # Signal plotting logic
    ├── run_study.py        # Execute backtest on chosen signal definitions
    └── symbol_data.py      # Symbol-level data abstraction

./classical_trading/        # Classical technical studies
    ├── custom_studies/
        └── rsi_macd_cvd_chai_custom_study.py
    └── indicators/
        ├── base_indicators_tos.py
        └── custom_indicators.py
        
./Parameter optimization on created study/        
    ├── optimization/
        └── run_optimization.py
        └── strat_optimizer.py


./docs/
    └── Example Signals and Indicators.png

📌 Getting Started

1️⃣ Install Dependencies

pip install -r requirements.txt

2️⃣ Update the Data

python UpdateMentionData.py

This will fetch and preprocess new mention and stock price data from QuiverQuant, Yahoo Finance, and other sources.

3️⃣ Run the Hyperparameter Search

from event_study.event_studies import hyperparam_search

df_results = hyperparam_search(df, ticker='AAPL')

4️⃣ View the Best Event Signals

• Check results: The system will output a ranked table of mention spike configurations with the highest predictive power.
• Visualize: Use utils/plotting.py to generate plots of post-event return trajectories.

🔍 Future Work

• Expand analysis to multiple tickers
• Integrate alternative sentiment sources (Twitter, Reddit, etc.)
• Develop real-time signal monitoring for trading execution
• Implement backtesting framework for strategy validation

📌 Contributions

If you have ideas to enhance the strategy, feel free to open a pull request! 🚀