🚀 RocketLabStockPredictor
A machine learning project that predicts the next-day direction of Rocket Lab (RKLB) stock using technical indicators, financial news sentiment analysis, and multiple machine learning models.
📈 Project Overview
The goal of this project is to investigate whether stock price movements can be predicted using:
Technical indicators Market context Trading volume Financial news sentiment Machine learning models
The project uses Rocket Lab (RKLB) as the target stock and compares its performance against the broader market.
🛠 Technologies Used Data Sources Yahoo Finance (historical stock data) Finnhub API (financial news) Machine Learning Logistic Regression Random Forest NLP FinBERT (ProsusAI/FinBERT) Python Libraries pandas numpy scikit-learn yfinance transformers matplotlib 📊 Features Technical Indicators RSI (Relative Strength Index) Moving Average (MA30) Exponential Moving Average (EMA30) Volatility Price vs EMA Volume Features Volume Moving Average Volume Ratio Volume Trend Features Market Features SPY Daily Returns Market Outperformance News Sentiment Features
Financial news headlines and summaries are analyzed using FinBERT.
Pipeline:
Financial News → FinBERT → Daily Sentiment Score → Machine Learning Model
🤖 Models Tested Logistic Regression
Used as a baseline model for binary classification.
Random Forest
Used to capture non-linear relationships between features.
📈 Evaluation Metrics
The project evaluates model performance using:
Accuracy Precision Recall F1 Score ROC-AUC Cross Validation 🔍 Key Findings Technical indicators provide useful predictive signals. Volume-based features showed measurable influence. Financial news sentiment contributed modestly to model performance. Predicting next-day stock direction remains extremely challenging. Random Forest slightly outperformed Logistic Regression. 📂 Project Structure RocketLabStockPredictor/ │ ├── news_downloader.py ├── news_processing.py ├── stock_analyzer_v2.py ├── README.md 🚀 Future Improvements QQQ and IWM market features Deep Learning models Social media sentiment analysis Hype Stock Finder integration Portfolio-level predictions Disclaimer
This project was created for educational and portfolio purposes only.
It does not constitute financial advice.