BSE SENSEX Simulation Engine with News Sentiment Analysis

This project simulates the BSE SENSEX market movements by incorporating real-world market data and news sentiment analysis. It fetches historical BSE SENSEX data, analyzes financial news sentiment, and simulates trading strategies based on market conditions and sentiment scores.

Features

• Real-time Data Fetching: Pulls historical BSE SENSEX data using Yahoo Finance
• News Sentiment Analysis: Analyzes financial news articles for market sentiment
• Trading Simulation: Simulates trading strategies based on market data and sentiment
• Performance Metrics: Calculates key performance indicators like returns, volatility, and drawdown
• Visualization: Generates interactive plots of simulation results

Prerequisites

• Python 3.7+
• NewsAPI key (free tier available at NewsAPI)

Installation

1. Clone the repository:

   git clone <repository-url>
   cd sensex-simulation-engine

2. Create and activate a virtual environment (recommended):

   python -m venv venv
   source venv/bin/activate  # On Windows: venv\Scripts\activate

3. Install the required packages:

   pip install -r requirements.txt

4. Create a .env file in the project root and add your NewsAPI key:

   NEWS_API_KEY=your_news_api_key_here
   

Usage

Basic Usage

Run the simulation with default parameters:

python main.py

Command Line Arguments

• --initial: Initial investment amount in INR (default: 1,000,000)
• --days: Number of days to simulate (default: 30)
• --sentiment-impact: Impact factor of sentiment on prices (default: 0.05)
• --no-news: Run simulation without news sentiment analysis

Examples

Run a 90-day simulation with 2 million INR initial investment:

python main.py --days 90 --initial 2000000

Run without news sentiment analysis:

python main.py --no-news

Project Structure

• main.py: Main script to run the simulation
• data_fetcher.py: Fetches BSE SENSEX historical data
• news_fetcher.py: Fetches and processes financial news
• sentiment_analyzer.py: Analyzes sentiment of news articles
• simulation_engine.py: Core simulation logic
• config.py: Configuration settings and constants
• data/: Directory for storing data files
  • bse_sensex_historical.csv: Cached historical data
  • financial_news.json: Cached news articles
  • simulation_results.csv: Simulation results
  • simulation_results.html: Interactive visualization

Output

The simulation will generate:

1. A CSV file with detailed simulation results
2. An interactive HTML plot showing:
   • Portfolio value over time
   • SENSEX price movements
   • Sentiment scores
   • Buy/sell signals

Performance Metrics

The simulation calculates several performance metrics:

• Total Return
• Annualized Return
• Volatility
• Sharpe Ratio
• Maximum Drawdown
• Number of Trades

Notes

• The simulation uses a simple trading strategy based on sentiment scores. You can modify the _apply_trading_strategy method in simulation_engine.py to implement your own strategies.
• News sentiment analysis uses both TextBlob and VADER for more robust sentiment detection.
• The free tier of NewsAPI has rate limits. For extensive backtesting, consider upgrading to a paid plan.

License

This project is licensed under the MIT License - see the LICENSE file for details.

Acknowledgments

• Yahoo Finance for market data
• NewsAPI for news data
• NLTK and TextBlob for sentiment analysis
• Plotly for visualization