{"title":"使用动态模块化光谱算法进行纵向市场结构检测","authors":"Philipp Wirth, Francesca Medda, Thomas Schröder","doi":"arxiv-2407.04500","DOIUrl":null,"url":null,"abstract":"In this paper, we introduce the Dynamic Modularity-Spectral Algorithm\n(DynMSA), a novel approach to identify clusters of stocks with high\nintra-cluster correlations and low inter-cluster correlations by combining\nRandom Matrix Theory with modularity optimisation and spectral clustering. The\nprimary objective is to uncover hidden market structures and find diversifiers\nbased on return correlations, thereby achieving a more effective risk-reducing\nportfolio allocation. We applied DynMSA to constituents of the S&P 500 and\ncompared the results to sector- and market-based benchmarks. Besides the\nconception of this algorithm, our contributions further include implementing a\nsector-based calibration for modularity optimisation and a correlation-based\ndistance function for spectral clustering. Testing revealed that DynMSA\noutperforms baseline models in intra- and inter-cluster correlation\ndifferences, particularly over medium-term correlation look-backs. It also\nidentifies stable clusters and detects regime changes due to exogenous shocks,\nsuch as the COVID-19 pandemic. Portfolios constructed using our clusters showed\nhigher Sortino and Sharpe ratios, lower downside volatility, reduced maximum\ndrawdown and higher annualised returns compared to an equally weighted market\nbenchmark.","PeriodicalId":501045,"journal":{"name":"arXiv - QuantFin - Portfolio Management","volume":"57 1","pages":""},"PeriodicalIF":0.0000,"publicationDate":"2024-07-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Longitudinal market structure detection using a dynamic modularity-spectral algorithm\",\"authors\":\"Philipp Wirth, Francesca Medda, Thomas Schröder\",\"doi\":\"arxiv-2407.04500\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"In this paper, we introduce the Dynamic Modularity-Spectral Algorithm\\n(DynMSA), a novel approach to identify clusters of stocks with high\\nintra-cluster correlations and low inter-cluster correlations by combining\\nRandom Matrix Theory with modularity optimisation and spectral clustering. The\\nprimary objective is to uncover hidden market structures and find diversifiers\\nbased on return correlations, thereby achieving a more effective risk-reducing\\nportfolio allocation. We applied DynMSA to constituents of the S&P 500 and\\ncompared the results to sector- and market-based benchmarks. Besides the\\nconception of this algorithm, our contributions further include implementing a\\nsector-based calibration for modularity optimisation and a correlation-based\\ndistance function for spectral clustering. Testing revealed that DynMSA\\noutperforms baseline models in intra- and inter-cluster correlation\\ndifferences, particularly over medium-term correlation look-backs. It also\\nidentifies stable clusters and detects regime changes due to exogenous shocks,\\nsuch as the COVID-19 pandemic. Portfolios constructed using our clusters showed\\nhigher Sortino and Sharpe ratios, lower downside volatility, reduced maximum\\ndrawdown and higher annualised returns compared to an equally weighted market\\nbenchmark.\",\"PeriodicalId\":501045,\"journal\":{\"name\":\"arXiv - QuantFin - Portfolio Management\",\"volume\":\"57 1\",\"pages\":\"\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2024-07-05\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"arXiv - QuantFin - Portfolio Management\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/arxiv-2407.04500\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"arXiv - QuantFin - Portfolio Management","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/arxiv-2407.04500","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
Longitudinal market structure detection using a dynamic modularity-spectral algorithm
In this paper, we introduce the Dynamic Modularity-Spectral Algorithm
(DynMSA), a novel approach to identify clusters of stocks with high
intra-cluster correlations and low inter-cluster correlations by combining
Random Matrix Theory with modularity optimisation and spectral clustering. The
primary objective is to uncover hidden market structures and find diversifiers
based on return correlations, thereby achieving a more effective risk-reducing
portfolio allocation. We applied DynMSA to constituents of the S&P 500 and
compared the results to sector- and market-based benchmarks. Besides the
conception of this algorithm, our contributions further include implementing a
sector-based calibration for modularity optimisation and a correlation-based
distance function for spectral clustering. Testing revealed that DynMSA
outperforms baseline models in intra- and inter-cluster correlation
differences, particularly over medium-term correlation look-backs. It also
identifies stable clusters and detects regime changes due to exogenous shocks,
such as the COVID-19 pandemic. Portfolios constructed using our clusters showed
higher Sortino and Sharpe ratios, lower downside volatility, reduced maximum
drawdown and higher annualised returns compared to an equally weighted market
benchmark.