Skip to content

the-agentic-ai/bitcoin-price-forecasting-using-prophet

Repository files navigation

Enhanced Bitcoin Price Forecasting for 2024 Halving Using Prophet

This project leverages the Prophet time series forecasting model to predict Bitcoin prices, incorporating five years of comprehensive market data. The focus is on critical market dynamics such as volatility, market cycles, investor sentiment, and particularly the 2024 Bitcoin halving event—a pivotal occurrence historically influencing Bitcoin's value. By analyzing data around previous halvings, the model aims to uncover trends crucial for forecasting future Bitcoin prices.

Features and Benefits

  • Prophet Forecasting Model: Utilizes Facebook's Prophet model for its intuitive and robust time series forecasting capabilities.
  • Halving Events Analysis: Examines the impact of Bitcoin halving events on market price dynamics.
  • Comprehensive Data Analysis: Includes an analysis of volatility, market cycles, and investor sentiment to enhance forecasting accuracy.
  • Visualization: Offers clear visualizations for comparing actual prices against forecasted scenarios.

Methodology

The methodology employs the Prophet model to analyze historical Bitcoin data. Special attention is given to periods surrounding Bitcoin halving events to enhance the accuracy of future price predictions.

Disclaimer and Warning

  • Educational Purpose Only: This research is solely for educational purposes and constitutes a self-study project.
  • Not Financial Advice: Predictions generated by this code are for educational use only and should not be considered financial or investment advice.
  • No Responsibility: The author and contributors to this repository assume no responsibility for any financial losses incurred.
  • Accuracy Not Guaranteed: The predictive performance of these models cannot be guaranteed.
  • Independent research: Users are encouraged to research and consult with professionals.
  • Compliance with Laws: Users must ensure compliance with all applicable laws and regulations in their jurisdiction.

Prerequisites

  • Python 3.11
  • Git

Installation

Setting Up a Virtual Environment

First, create and activate a virtual environment:

python -m venv .venv  # Create a virtual environment
source .venv/bin/activate  # Activate on macOS and Linux
.venv\Scripts\activate  # Activate on Windows

Installing Dependencies

Install the required dependencies:

pip install -r requirements.txt  # Install Python dependencies
npm install dukascopy-node --save  # Install Node.js dependencies

Data Collection

chmod +x data-download.sh
./data-download.sh

Quick Start

Windows Specific Notes

  • Use PowerShell or Git Bash for shell commands.
  • Ensure Docker Desktop for Windows is set to Linux containers.
  • Python commands may require using py instead of python or python3.

Clone the Repository

To get started, clone the repository:

git clone git@github.com:SMARTSHEEP-IO/bitcoin-price-forecasting-using-prophet.git

Usage

Execute the main.py script to start the data processing and model training/prediction process:

python main.py

Support and Subscribe

Contributing

Contributions to improve the project are welcome. Please adhere to standard open-source contribution guidelines.

Citation

If this project aids in your research or any other project, kindly cite it as follows:

    @misc{Samizadeh2024BitcoinProphet,
       author = {Samizadeh, Iman},
       title = {{Enhanced Bitcoin Price Forecasting for 2024 Halving Using Prophet}},
       year = {2024},
       howpublished = {\url{https://github.com/SMARTSHEEP-IO/bitcoin-price-forecasting-using-prophet}}
    }

Credit:

  1. https://github.com/Leo4815162342/dukascopy-node
  2. Prophet Documentation: https://facebook.github.io/prophet/

Keywords: Bitcoin, Halving, Prediction, Volatility, Growth, ATH, Cryptocurrency, Analysis, Financial Modeling, Prophet

About

Forecasting the price of Bitcoin’s during or post 2024 halving using Facebook Prophet

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published