Technology Category
Data Mining & Knowledge Management: Graph Mining, Social Network Analysis & CommunityApplication Domains: Humanities & Computational Social ScienceSearch and Optimization: Distributed Search
Network-based diversification of stock and cryptocurrency portfolios
🤖 AI Summary
This paper addresses the network-based diversification problem in cross-market portfolios integrating equities and cryptocurrencies. Methodologically, it proposes an asset correlation network model that jointly incorporates Pearson correlation and mutual information distance; dynamically identifies multi-scale communities using integrated Louvain and Affinity Propagation (AP) clustering; and constructs diversified portfolios via centrality-based selection, PCA-driven variance contribution weighting, and maximum spanning tree optimization. Its key contributions include: (i) the first systematic empirical revelation of structural divergence between equity and crypto markets during the COVID-19 pandemic and the Ukraine crisis—equity communities exhibit low intra-community correlation, whereas cryptocurrency markets display inverse clustering; and (ii) a novel monthly co-occurrence network framework coupled with a community-driven stock selection strategy. Empirical results demonstrate that the proposed portfolio significantly reduces volatility and achieves a higher Sharpe ratio than the equal-weighted benchmark.
Technology Category
Application Category
📝 Abstract
Maintaining a balance between returns and volatility is a common strategy for portfolio diversification, whether investing in traditional equities or digital assets like cryptocurrencies. One approach for diversification is the application of community detection or clustering, using a network representing the relationships between assets. We examine two network representations, one based on a standard distance matrix based on correlation, and another based on mutual information. The Louvain and Affinity propagation algorithms were employed for finding the network communities (clusters) based on annual data. Furthermore, we examine building assets' co-occurrence networks, where communities are detected for each month throughout a whole year and then the links represent how often assets belong to the same community. Portfolios are then constructed by selecting several assets from each community based on local properties (degree centrality), global properties (closeness centrality), or explained variance (Principal component analysis), with three value ranges (max, med, min), calculated on a maximal spanning tree or a fully connected community sub-graph. We explored these various strategies on data from the S&P 500 and the Top 203 cryptocurrencies with a market cap above 2M USD in the period from Jan 2019 to Sep 2022. Moreover, we study into more details the periods of the beginning of the COVID-19 outbreak and the start of the war in Ukraine. The results confirm some of the previous findings already known for traditional stock markets and provide some further insights, while they reveal an opposing trend in the crypto-assets market.
Problem
Research questions and friction points this paper is trying to address.
Compare network methods for stock and crypto portfolio diversification
Analyze community detection impact on asset return and volatility
Evaluate portfolio strategies during COVID-19 and Ukraine war crises
Innovation
Methods, ideas, or system contributions that make the work stand out.
Using correlation and mutual information networks
Applying Louvain and Affinity propagation algorithms
Constructing portfolios with centrality and PCA
🔎 Similar Papers
No similar papers found.
