# A network-based strategy of price correlations for optimal cryptocurrency portfolios

**Authors:** Ruixue Jing, Luis Enrique Correa Rocha  
**arXiv:** [2304.02362](https://arxiv.org/abs/2304.02362)  
**Date:** 2023-04-05  
**Summary page:** https://www.summarizepaper.com/en/paper/2304.02362v1/

## Key points

- Cryptocurrencies have emerged as a new asset class in finance
- Investors are exploring portfolio profitability potential instead of single coins
- Price forecasting is challenging due to market volatility, making portfolios particularly useful
- The crypto market is a self-organized complex system with inter-dependencies between cryptocurrencies that can be exploited for understanding market dynamics and building efficient portfolios
- Ruixue Jing and Luis Enrique Correa Rocha use network methods to identify highly decorrelated cryptocurrencies for diversified portfolios using the Markowitz Portfolio Theory agnostic to future market behavior
- Past price correlations reduce risk and may improve performance compared to methodologies based exclusively on price auto-correlations
- Network-based portfolios with 46 coins perform optimally and outperform benchmarks up to an investment horizon of 14 days, reaching up to 1,066% average expected return within one day with reasonable associated risks
- Popular cryptocurrencies are typically not included in optimal portfolios
- Short-term crypto investments may be competitive with traditional high-risk investments but caution is advised due to high variability of prices.

## Summary

In the world of finance, cryptocurrencies have emerged as a new asset class that is maintained by a decentralized system using cryptography. Investors in this emerging digital market are exploring the profitability potential of portfolios in place of single coins. However, price forecasting in such a volatile market is challenging, and hence portfolios are particularly useful. The crypto market is a self-organized complex system where the complex inter-dependencies between cryptocurrencies may be exploited to understand the market dynamics and build efficient portfolios. In this letter titled "A network-based strategy of price correlations for optimal cryptocurrency portfolios," Ruixue Jing and Luis Enrique Correa Rocha use network methods to identify highly decorrelated cryptocurrencies to create diversified portfolios using the Markowitz Portfolio Theory agnostic to future market behavior. The authors show that past price correlations reduce risk and may improve the performance of crypto portfolios compared to methodologies based exclusively on price auto-correlations. The performance of their network-based portfolios is optimal with 46 coins and superior to benchmarks up to an investment horizon of 14 days, reaching up to 1,066% average expected return within one day with reasonable associated risks. Interestingly, popular cryptocurrencies are typically not included in these optimal portfolios. The authors caution that short-term crypto investments may be competitive with traditional high-risk investments such as the stock or commodity markets but call for caution given the high variability of prices. Overall, this study provides insights into how investors can exploit complex interdependencies between cryptocurrencies to build efficient portfolios that offer higher returns with lower risks than traditional approaches.