{"title":"DSPO:直接分类投资组合构建的端到端框架","authors":"Jianyuan Zhong, Zhijian Xu, Saizhuo Wang, Xiangyu Wen, Jian Guo, Qiang Xu","doi":"arxiv-2405.15833","DOIUrl":null,"url":null,"abstract":"In quantitative investment, constructing characteristic-sorted portfolios is\na crucial strategy for asset allocation. Traditional methods transform raw\nstock data of varying frequencies into predictive characteristic factors for\nasset sorting, often requiring extensive manual design and misalignment between\nprediction and optimization goals. To address these challenges, we introduce\nDirect Sorted Portfolio Optimization (DSPO), an innovative end-to-end framework\nthat efficiently processes raw stock data to construct sorted portfolios\ndirectly. DSPO's neural network architecture seamlessly transitions stock data\nfrom input to output while effectively modeling the intra-dependency of\ntime-steps and inter-dependency among all tradable stocks. Additionally, we\nincorporate a novel Monotonical Logistic Regression loss, which directly\nmaximizes the likelihood of constructing optimal sorted portfolios. To the best\nof our knowledge, DSPO is the first method capable of handling market\ncross-sections with thousands of tradable stocks fully end-to-end from raw\nmulti-frequency data. Empirical results demonstrate DSPO's effectiveness,\nyielding a RankIC of 10.12\\% and an accumulated return of 121.94\\% on the New\nYork Stock Exchange in 2023-2024, and a RankIC of 9.11\\% with a return of\n108.74\\% in other markets during 2021-2022.","PeriodicalId":501045,"journal":{"name":"arXiv - QuantFin - Portfolio Management","volume":"17 1","pages":""},"PeriodicalIF":0.0000,"publicationDate":"2024-05-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"DSPO: An End-to-End Framework for Direct Sorted Portfolio Construction\",\"authors\":\"Jianyuan Zhong, Zhijian Xu, Saizhuo Wang, Xiangyu Wen, Jian Guo, Qiang Xu\",\"doi\":\"arxiv-2405.15833\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"In quantitative investment, constructing characteristic-sorted portfolios is\\na crucial strategy for asset allocation. Traditional methods transform raw\\nstock data of varying frequencies into predictive characteristic factors for\\nasset sorting, often requiring extensive manual design and misalignment between\\nprediction and optimization goals. To address these challenges, we introduce\\nDirect Sorted Portfolio Optimization (DSPO), an innovative end-to-end framework\\nthat efficiently processes raw stock data to construct sorted portfolios\\ndirectly. DSPO's neural network architecture seamlessly transitions stock data\\nfrom input to output while effectively modeling the intra-dependency of\\ntime-steps and inter-dependency among all tradable stocks. Additionally, we\\nincorporate a novel Monotonical Logistic Regression loss, which directly\\nmaximizes the likelihood of constructing optimal sorted portfolios. To the best\\nof our knowledge, DSPO is the first method capable of handling market\\ncross-sections with thousands of tradable stocks fully end-to-end from raw\\nmulti-frequency data. Empirical results demonstrate DSPO's effectiveness,\\nyielding a RankIC of 10.12\\\\% and an accumulated return of 121.94\\\\% on the New\\nYork Stock Exchange in 2023-2024, and a RankIC of 9.11\\\\% with a return of\\n108.74\\\\% in other markets during 2021-2022.\",\"PeriodicalId\":501045,\"journal\":{\"name\":\"arXiv - QuantFin - Portfolio Management\",\"volume\":\"17 1\",\"pages\":\"\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2024-05-24\",\"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-2405.15833\",\"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-2405.15833","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
DSPO: An End-to-End Framework for Direct Sorted Portfolio Construction
In quantitative investment, constructing characteristic-sorted portfolios is
a crucial strategy for asset allocation. Traditional methods transform raw
stock data of varying frequencies into predictive characteristic factors for
asset sorting, often requiring extensive manual design and misalignment between
prediction and optimization goals. To address these challenges, we introduce
Direct Sorted Portfolio Optimization (DSPO), an innovative end-to-end framework
that efficiently processes raw stock data to construct sorted portfolios
directly. DSPO's neural network architecture seamlessly transitions stock data
from input to output while effectively modeling the intra-dependency of
time-steps and inter-dependency among all tradable stocks. Additionally, we
incorporate a novel Monotonical Logistic Regression loss, which directly
maximizes the likelihood of constructing optimal sorted portfolios. To the best
of our knowledge, DSPO is the first method capable of handling market
cross-sections with thousands of tradable stocks fully end-to-end from raw
multi-frequency data. Empirical results demonstrate DSPO's effectiveness,
yielding a RankIC of 10.12\% and an accumulated return of 121.94\% on the New
York Stock Exchange in 2023-2024, and a RankIC of 9.11\% with a return of
108.74\% in other markets during 2021-2022.