Siudak Dariusz
Division of Economics and Finance, Institute of Management, Lodz University of Technology, Lodz, Poland.
PLoS One. 2025 Apr 18;20(4):e0320799. doi: 10.1371/journal.pone.0320799. eCollection 2025.
This work describes the design of a novel financial multiplex network composed of three layers obtained by applying the MST-based cross-correlation network, using the data from 465 companies listed on the US market. The study employs a combined approach of complex multiplex networks, to examine the statistical properties of asset interdependence within the financial market. In addition, it performs an extensive analysis of both the similarities and the differences between this financial multiplex network, its individual layers, and the commonly studied stock return network. The results highlight the importance of the financial multiplex network, demonstrating that its network layers offer unique information within the multiplex dataset. Empirical analysis reveals dissimilarities between the financial multiplex network and the stock return monoplex network, indicating that the two networks provide distinct insights into the structure of the stock market. Furthermore, the financial multiplex network outperforms the singleplex network of stock returns because it has a structure that better determines the future Sharpe ratio. These findings add substantially to our understanding of the financial market system in which multiple types of relationship among financial assets play an important role.
这项工作描述了一种新型金融多重网络的设计，该网络由三层组成，通过应用基于最小生成树的互相关网络获得，使用的是美国市场上465家上市公司的数据。该研究采用了复杂多重网络的组合方法，以检验金融市场内资产相互依存的统计特性。此外，它还对这个金融多重网络、其各个层以及常用的股票收益网络之间的异同进行了广泛分析。结果突出了金融多重网络的重要性，表明其网络层在多重数据集中提供了独特信息。实证分析揭示了金融多重网络与股票收益单重网络之间的差异，表明这两个网络对股票市场结构提供了不同的见解。此外，金融多重网络优于股票收益单重网络，因为它具有能更好地确定未来夏普比率的结构。这些发现极大地增进了我们对金融市场系统的理解，在该系统中金融资产之间的多种关系起着重要作用。
