Balancing cryptoassets and commodities: novel Weighted-Risk-Contribution indices for the alternative asset space
File(s)
Author(s)
Koutsouri, Katerina
Type
Thesis
Abstract
This thesis aims to serve as a comprehensive guide for effective design, development and performance evaluation of novel Weighted-Risk-Contribution indices for the alternative asset space. We concentrate our focus on the quest for diversification and risk-reward balance by incorporating commodities into a crypto-asset allocation.
First, we introduce the CoinShares Gold and Cryptoassets Index, a novel index product that combines a basket of five cryptoassets with gold to enhance the risk profile while maintaining independence from traditional financial asset classes. By generalizing the theory of Equal Risk Contribution, we compare various asset allocation strategies and demonstrate the effectiveness of a crypto-gold weighting based on the Weighted Risk Contribution allocation scheme in terms of Sharpe Ratio.
To assess the resilience of the index, we further introduce a complete stress testing framework using ARMA--GARCH processes and copulas to simulate realistic market conditions and extreme events. The analysis reveals a superior risk-return profile for the CoinShares Gold and Cryptoassets Index compared to traditional market-cap-weighted cryptoasset indices. Furthermore, we employ Gaussian Hidden Markov Models and Markov-switching GARCH models to identify high-risk market conditions and demonstrate the stable risk-reward profile and superior performance of the index in terms of the Omega ratio, particularly for investors targeting wealth preservation and moderate annual returns.
Lastly, we seek to quantify the diversification benefits of incorporating commodities into cryptoasset portfolios by comparing the CoinShares Gold and Cryptoassets Index with a modified index that replaces gold with a basket of five commodities. Mean-variance spanning tests and simulation-based Dynamic Conditional Correlation GARCH models reveal statistically significant improvements in the efficient frontier for both indices. We conclude that the modified index is more suitable for investors seeking higher annual returns, while the original index is more appropriate for those with moderate annual return goals.
The aforementioned studies advance our understanding of portfolio diversification in the context of cryptoassets and emphasize the potential benefits of incorporating gold and other commodities into crypto-based index strategies, thereby providing valuable insights for investors and financial practitioners.
First, we introduce the CoinShares Gold and Cryptoassets Index, a novel index product that combines a basket of five cryptoassets with gold to enhance the risk profile while maintaining independence from traditional financial asset classes. By generalizing the theory of Equal Risk Contribution, we compare various asset allocation strategies and demonstrate the effectiveness of a crypto-gold weighting based on the Weighted Risk Contribution allocation scheme in terms of Sharpe Ratio.
To assess the resilience of the index, we further introduce a complete stress testing framework using ARMA--GARCH processes and copulas to simulate realistic market conditions and extreme events. The analysis reveals a superior risk-return profile for the CoinShares Gold and Cryptoassets Index compared to traditional market-cap-weighted cryptoasset indices. Furthermore, we employ Gaussian Hidden Markov Models and Markov-switching GARCH models to identify high-risk market conditions and demonstrate the stable risk-reward profile and superior performance of the index in terms of the Omega ratio, particularly for investors targeting wealth preservation and moderate annual returns.
Lastly, we seek to quantify the diversification benefits of incorporating commodities into cryptoasset portfolios by comparing the CoinShares Gold and Cryptoassets Index with a modified index that replaces gold with a basket of five commodities. Mean-variance spanning tests and simulation-based Dynamic Conditional Correlation GARCH models reveal statistically significant improvements in the efficient frontier for both indices. We conclude that the modified index is more suitable for investors seeking higher annual returns, while the original index is more appropriate for those with moderate annual return goals.
The aforementioned studies advance our understanding of portfolio diversification in the context of cryptoassets and emphasize the potential benefits of incorporating gold and other commodities into crypto-based index strategies, thereby providing valuable insights for investors and financial practitioners.
Version
Open Access
Date Issued
2023-04
Date Awarded
2024-02
Copyright Statement
Creative Commons Attribution NonCommercial Licence
License URL
Advisor
Knottenbelt, William J.
Publisher Department
Computing
Publisher Institution
Imperial College London
Qualification Level
Doctoral
Qualification Name
Doctor of Philosophy (PhD)