This is a project of a dissertation competition from an bachelor degree of economics student from Soochow University. Topic: Predicting Ethereum returns by using time series
This study employs autoregressive and cross-period regression models from time series analysis, using various cryptocurrency prices as variables to predict the future returns of Ethereum. We find that platform tokens are the most suitable variables for predicting Ethereum returns. Additionally, we discover that NFT-related digital currencies, which are traded on the Ethereum blockchain, can also predict Ethereum returns. Conversely, DeFi-related tokens and meme coins are the least suitable for predicting Ethereum returns.
-
Data Collection:
- Historical price data for Ethereum and other cryptocurrencies are collected.
-
Data Preprocessing:
- Cleaning and preparing the data for analysis.
-
Model Building:
- Using autoregressive and cross-period regression models to predict Ethereum returns.
-
Analysis:
- Evaluating the predictive power of different types of cryptocurrencies.
- Platform tokens are the most suitable variables for predicting Ethereum returns.
- NFT-related digital currencies traded on the Ethereum blockchain also have predictive power.
- DeFi-related tokens and meme coins are the least suitable for predicting Ethereum returns.
- Python 3.7+
- Jupyter Notebook
- pandas
- numpy
- statsmodels
- matplotlib
To run the analysis, open the Dissertation_code.ipynb notebook in Jupyter and follow the steps provided. Ensure you have the necessary data files in the data/ directory.
Contributions are welcome! Please feel free to submit a Pull Request.
This project is licensed under the MIT License.
GitHub: Chiang0111
G-Mail: [email protected]