Time Series Modeling Project

This repository contains code for a time series modeling project that explores the application of three different models: Artificial Neural Networks (ANN), Autoregressive Integrated Moving Average (ARIMA), and Long Short-Term Memory (LSTM) networks.

Project Overview

The goal of this project is to develop and compare the performance of various time series forecasting models on a given dataset of Microsoft (MSFT) stock closing prices. Time series forecasting is a critical task in many domains such as finance, weather prediction, and sales forecasting. By experimenting with different models, we aim to identify the most suitable approach for forecasting MSFT stock prices.

Models Implemented

1. Artificial Neural Networks (ANN)

• Description: ANN is a type of machine learning model inspired by the biological neural networks of the human brain. It consists of interconnected nodes (neurons) organized in layers. In this project, we utilize ANN for time series forecasting.

2. Autoregressive Integrated Moving Average (ARIMA)

• Description: ARIMA is a classical statistical model commonly used for time series forecasting. It models the next step in the sequence as a linear function of the observations and residual errors at prior time steps.

3. Long Short-Term Memory (LSTM)

• Description: LSTM is a type of recurrent neural network (RNN) architecture designed to capture long-term dependencies in sequence data. It is well-suited for modeling time series data due to its ability to retain information over long periods.

Dataset

The dataset used in this project consists of historical daily closing prices of Microsoft (MSFT) stock. Each observation represents the closing price of MSFT stock at the end of a trading day.

Tools and Technologies Used

• Python
• TensorFlow/Keras (for ANN and LSTM)
• Statsmodels (for ARIMA)
• Pandas (for data manipulation)
• Matplotlib/Seaborn (for visualization)
• GitHub (for version control and collaboration)

Usage

To use the code in this repository, follow these steps:

1. Clone the repository to your local machine:

git clone https://github.com/MouadAH2003/TS_projects.git

2. Navigate to the cloned repository:

cd TS_projects

3. Install the required dependencies using pip:

pip install -r requirements.txt

4. launch jupyter notebook to execute the time series modeling and forecasting:

jupyter notebook

5. Explore the results and visualizations generated by the models.

Results

After running the code, you will find the following results:

• Evaluation metrics: Accuracy, Mean Squared Error (MSE), etc.
• Visualizations: Plots showing the predicted vs. actual stock prices for each model.
• Insights: Analysis and interpretations of the model performance and forecasted trends.

License

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

Acknowledgements

• This project was inspired by various GitHub repositories, where code examples and implementations provided valuable insights into time series modeling techniques.
• Google searches and online resources played a significant role in discovering relevant literature, tutorials, and discussions related to time series forecasting, contributing to the project's development.
• Special thanks to private sources, including internal documentation, discussions, and insights from colleagues, which enriched the understanding and implementation of time series models in this project.

---