Gal Oblak
Context:
Stock market prices are influenced by a multitude of factors, including financial indicators, market sentiment, macroeconomic data, and global events. Predicting stock prices can help investors make informed decisions and optimize their trading strategies.
My Contribution:
I built a machine learning model to predict stock market prices based on historical data and various economic indicators. The project comprised several essential stages:
Collected historical stock price data, financial indicators (e.g., moving averages, volume, volatility), and macroeconomic variables (e.g., interest rates, GDP growth).
Preprocessed the data by normalizing features, handling missing values, and creating time series sequences.
Conducted EDA to identify trends, patterns, and correlations within the data.
Visualized stock price movements and significant economic indicators using tools like matplotlib and seaborn.
Trained multiple regression models, including linear regression, support vector machines, and advanced deep learning models like Long Short-Term Memory (LSTM) networks.
Evaluated model performance using metrics such as Mean Absolute Error (MAE) and Root Mean Squared Error (RMSE).
Fine-tuned the models through hyperparameter optimization and cross-validation techniques.
Analyzed feature importance to understand the most significant factors influencing stock price movements.
Utilized techniques such as SHAP values to interpret model predictions and ensure transparency.
Implemented real-time data updates to provide the latest stock price forecasts.
This project successfully demonstrated the ability of machine learning models to predict stock market prices, offering valuable insights for investors and financial analysts.