This project explored time series forecasting applied to stock price prediction using historical data from Yahoo Finance. Working in Google Colab, I implemented two forecasting approaches: Autoregressive Integrated Moving Average (ARIMA) for classical statistical modeling, and Long Short-Term Memory (LSTM) networks for deep learning-based sequence prediction.
The final model focused on forecasting Google’s stock price and achieved approximately 86% accuracy. The project deepened my understanding of the trade-offs between traditional statistical methods and neural network approaches for sequential data problems.
Here is a link to the project on Github.