Crypto Token Correlation Heatmap

This project analyzes and visualizes the correlation between multiple cryptocurrency tokens using historical price data from the CoinGecko Pro API.

Features

• Fetch historical price data for any set of tokens
• Calculate log returns and token volatility
• Generate a correlation heatmap for selected tokens
• Save and load processed data for fast analysis

Setup

1. Clone the repository

2. Install dependencies

   pip install -r requirements.txt
   

3. Add your CoinGecko Pro API key

   • Create a .env file in the token_heatmap folder:

     GECKO_API=your_api_key_here
     

Usage

1. Fetch and save token price data

   • Run 01_data.ipynb to fetch historical prices for your chosen tokens and save the DataFrame.

2. Analyze correlations

   • Open token_correlation.ipynb to load the saved data, calculate log returns, volatility, and plot the correlation heatmap.

Example

# Fetch prices for Bitcoin, Ethereum, and Solana
token_ids = ["bitcoin", "ethereum", "solana"]
api_key = os.getenv("GECKO_API")
df_prices = fetch_token_prices(token_ids, api_key=api_key)

# Plot correlation heatmap
corr = df_prices.pct_change().apply(lambda x: np.log1p(x)).corr()
sns.heatmap(corr, annot=True, cmap='coolwarm', vmin=-1, vmax=1)
plt.title("Token Correlation Heatmap")
plt.show()

Requirements

• Python 3.8+
• requests
• pandas
• numpy
• matplotlib
• seaborn
• python-dotenv
• joblib

License

MIT License

---

**Feel free to fork :)