AAPL Stock Price Forecasting

This project aims to forecast Apple (AAPL) stock prices using historical data from the last 5 years, retrieved from Yahoo Finance.
We implemented two approaches to build predictive models:

• Time Series Forecasting Method: Holt-Winters Exponential Smoothing for capturing level, trend, and seasonality.
• Deep Learning Method: Convolutional Neural Network (CNN) applied to stock price sequences for pattern recognition and future prediction.

---

Project Structure

• Data Preprocessing: Handling missing values, normalization, and technical indicator calculation (moving average, RSI).

• Time Series Model: Holt-Winters exponential smoothing for forecasting.

• Deep Learning Model: CNN architecture trained on sliding windows of stock prices.

• Evaluation: Performance metrics such as MSE, MAE, and RMSE are computed.

• Visualization: Historical vs. predicted stock prices plotted for analysis.

• Results

Holt-Winters Time Series Forecast

CNN Training Progress

CNN Validation & Forecast Results