arXiv - QuantFin - Portfolio Management最新文献

{"title":"Finding Near-Optimal Portfolios With Quality-Diversity","authors":"Bruno Gašperov, Marko Đurasević, Domagoj Jakobovic","doi":"arxiv-2402.16118","DOIUrl":"https://doi.org/arxiv-2402.16118","url":null,"abstract":"The majority of standard approaches to financial portfolio optimization (PO)\u0000are based on the mean-variance (MV) framework. Given a risk aversion\u0000coefficient, the MV procedure yields a single portfolio that represents the\u0000optimal trade-off between risk and return. However, the resulting optimal\u0000portfolio is known to be highly sensitive to the input parameters, i.e., the\u0000estimates of the return covariance matrix and the mean return vector. It has\u0000been shown that a more robust and flexible alternative lies in determining the\u0000entire region of near-optimal portfolios. In this paper, we present a novel\u0000approach for finding a diverse set of such portfolios based on\u0000quality-diversity (QD) optimization. More specifically, we employ the\u0000CVT-MAP-Elites algorithm, which is scalable to high-dimensional settings with\u0000potentially hundreds of behavioral descriptors and/or assets. The results\u0000highlight the promising features of QD as a novel tool in PO.","PeriodicalId":501045,"journal":{"name":"arXiv - QuantFin - Portfolio Management","volume":"102 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-02-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139977511","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Sizing the bets in a focused portfolio","authors":"Vuko Vukcevic, Robert Keser","doi":"arxiv-2402.15588","DOIUrl":"https://doi.org/arxiv-2402.15588","url":null,"abstract":"The paper provides a mathematical model and a tool for the focused investing\u0000strategy as advocated by Buffett, Munger, and others from this investment\u0000community. The approach presented here assumes that the investor's role is to\u0000think about probabilities of different outcomes for a set of businesses. Based\u0000on these assumptions, the tool calculates the optimal allocation of capital for\u0000each of the investment candidates. The model is based on a generalized Kelly\u0000Criterion with options to provide constraints that ensure: no shorting, limited\u0000use of leverage, providing a maximum limit to the risk of permanent loss of\u0000capital, and maximum individual allocation. The software is applied to an\u0000example portfolio from which certain observations about excessive\u0000diversification are obtained. In addition, the software is made available for\u0000public use.","PeriodicalId":501045,"journal":{"name":"arXiv - QuantFin - Portfolio Management","volume":"18 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-02-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139981435","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Combining Transformer based Deep Reinforcement Learning with Black-Litterman Model for Portfolio Optimization","authors":"Ruoyu SunXi'an Jiaotong-Liverpool University, School of Mathematics and Physics, Department of Financial and Actuarial Mathematics, Angelos StefanidisXi'an Jiaotong-Liverpool University Entrepreneur College, Zhengyong JiangXi'an Jiaotong-Liverpool University Entrepreneur College, Jionglong SuXi'an Jiaotong-Liverpool University Entrepreneur College","doi":"arxiv-2402.16609","DOIUrl":"https://doi.org/arxiv-2402.16609","url":null,"abstract":"As a model-free algorithm, deep reinforcement learning (DRL) agent learns and\u0000makes decisions by interacting with the environment in an unsupervised way. In\u0000recent years, DRL algorithms have been widely applied by scholars for portfolio\u0000optimization in consecutive trading periods, since the DRL agent can\u0000dynamically adapt to market changes and does not rely on the specification of\u0000the joint dynamics across the assets. However, typical DRL agents for portfolio\u0000optimization cannot learn a policy that is aware of the dynamic correlation\u0000between portfolio asset returns. Since the dynamic correlations among portfolio\u0000assets are crucial in optimizing the portfolio, the lack of such knowledge\u0000makes it difficult for the DRL agent to maximize the return per unit of risk,\u0000especially when the target market permits short selling (i.e., the US stock\u0000market). In this research, we propose a hybrid portfolio optimization model\u0000combining the DRL agent and the Black-Litterman (BL) model to enable the DRL\u0000agent to learn the dynamic correlation between the portfolio asset returns and\u0000implement an efficacious long/short strategy based on the correlation.\u0000Essentially, the DRL agent is trained to learn the policy to apply the BL model\u0000to determine the target portfolio weights. To test our DRL agent, we construct\u0000the portfolio based on all the Dow Jones Industrial Average constitute stocks.\u0000Empirical results of the experiments conducted on real-world United States\u0000stock market data demonstrate that our DRL agent significantly outperforms\u0000various comparison portfolio choice strategies and alternative DRL frameworks\u0000by at least 42% in terms of accumulated return. In terms of the return per unit\u0000of risk, our DRL agent significantly outperforms various comparative portfolio\u0000choice strategies and alternative strategies based on other machine learning\u0000frameworks.","PeriodicalId":501045,"journal":{"name":"arXiv - QuantFin - Portfolio Management","volume":"40 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-02-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139977276","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Randomized Control in Performance Analysis and Empirical Asset Pricing","authors":"Cyril Bachelard, Apostolos Chalkis, Vissarion Fisikopoulos, Elias Tsigaridas","doi":"arxiv-2403.00009","DOIUrl":"https://doi.org/arxiv-2403.00009","url":null,"abstract":"The present article explores the application of randomized control techniques\u0000in empirical asset pricing and performance evaluation. It introduces geometric\u0000random walks, a class of Markov chain Monte Carlo methods, to construct\u0000flexible control groups in the form of random portfolios adhering to investor\u0000constraints. The sampling-based methods enable an exploration of the\u0000relationship between academically studied factor premia and performance in a\u0000practical setting. In an empirical application, the study assesses the\u0000potential to capture premias associated with size, value, quality, and momentum\u0000within a strongly constrained setup, exemplified by the investor guidelines of\u0000the MSCI Diversified Multifactor index. Additionally, the article highlights\u0000issues with the more traditional use case of random portfolios for drawing\u0000inferences in performance evaluation, showcasing challenges related to the\u0000intricacies of high-dimensional geometry.","PeriodicalId":501045,"journal":{"name":"arXiv - QuantFin - Portfolio Management","volume":"319 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-02-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140026003","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Cyber risk and the cross-section of stock returns","authors":"Daniel Celeny, Loïc Maréchal","doi":"arxiv-2402.04775","DOIUrl":"https://doi.org/arxiv-2402.04775","url":null,"abstract":"We extract firms' cyber risk with a machine learning algorithm measuring the\u0000proximity between their disclosures and a dedicated cyber corpus. Our approach\u0000outperforms dictionary methods, uses full disclosure and not devoted-only\u0000sections, and generates a cyber risk measure uncorrelated with other firms'\u0000characteristics. We find that a portfolio of US-listed stocks in the high cyber\u0000risk quantile generates an excess return of 18.72% p.a. Moreover, a long-short\u0000cyber risk portfolio has a significant and positive risk premium of 6.93%\u0000p.a., robust to all factors' benchmarks. Finally, using a Bayesian asset\u0000pricing method, we show that our cyber risk factor is the essential feature\u0000that allows any multi-factor model to price the cross-section of stock returns.","PeriodicalId":501045,"journal":{"name":"arXiv - QuantFin - Portfolio Management","volume":"89 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-02-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139758466","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Developing A Multi-Agent and Self-Adaptive Framework with Deep Reinforcement Learning for Dynamic Portfolio Risk Management","authors":"Zhenglong Li, Vincent Tam, Kwan L. Yeung","doi":"arxiv-2402.00515","DOIUrl":"https://doi.org/arxiv-2402.00515","url":null,"abstract":"Deep or reinforcement learning (RL) approaches have been adapted as reactive\u0000agents to quickly learn and respond with new investment strategies for\u0000portfolio management under the highly turbulent financial market environments\u0000in recent years. In many cases, due to the very complex correlations among\u0000various financial sectors, and the fluctuating trends in different financial\u0000markets, a deep or reinforcement learning based agent can be biased in\u0000maximising the total returns of the newly formulated investment portfolio while\u0000neglecting its potential risks under the turmoil of various market conditions\u0000in the global or regional sectors. Accordingly, a multi-agent and self-adaptive\u0000framework namely the MASA is proposed in which a sophisticated multi-agent\u0000reinforcement learning (RL) approach is adopted through two cooperating and\u0000reactive agents to carefully and dynamically balance the trade-off between the\u0000overall portfolio returns and their potential risks. Besides, a very flexible\u0000and proactive agent as the market observer is integrated into the MASA\u0000framework to provide some additional information on the estimated market trends\u0000as valuable feedbacks for multi-agent RL approach to quickly adapt to the\u0000ever-changing market conditions. The obtained empirical results clearly reveal\u0000the potential strengths of our proposed MASA framework based on the multi-agent\u0000RL approach against many well-known RL-based approaches on the challenging data\u0000sets of the CSI 300, Dow Jones Industrial Average and S&P 500 indexes over the\u0000past 10 years. More importantly, our proposed MASA framework shed lights on\u0000many possible directions for future investigation.","PeriodicalId":501045,"journal":{"name":"arXiv - QuantFin - Portfolio Management","volume":"8 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139666253","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Optimal portfolio under ratio-type periodic evaluation in incomplete markets with stochastic factors","authors":"Wenyuan Wang, Kaixin Yan, Xiang Yu","doi":"arxiv-2401.14672","DOIUrl":"https://doi.org/arxiv-2401.14672","url":null,"abstract":"This paper studies a type of periodic utility maximization for portfolio\u0000management in an incomplete market model, where the underlying price diffusion\u0000process depends on some external stochastic factors. The portfolio performance\u0000is periodically evaluated on the relative ratio of two adjacent wealth levels\u0000over an infinite horizon. For both power and logarithmic utilities, we\u0000formulate the auxiliary one-period optimization problems with modified utility\u0000functions, for which we develop the martingale duality approach to establish\u0000the existence of the optimal portfolio processes and the dual minimizers can be\u0000identified as the \"least favorable\" completion of the market. With the help of\u0000the duality results in the auxiliary problems and some fixed point arguments,\u0000we further derive and verify the optimal portfolio processes in a periodic\u0000manner for the original periodic evaluation problems over an infinite horizon.","PeriodicalId":501045,"journal":{"name":"arXiv - QuantFin - Portfolio Management","volume":"23 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-01-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139582972","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Dynamic portfolio selection under generalized disappointment aversion","authors":"Zongxia Liang, Sheng Wang, Jianming Xia, Fengyi Yuan","doi":"arxiv-2401.08323","DOIUrl":"https://doi.org/arxiv-2401.08323","url":null,"abstract":"This paper addresses the continuous-time portfolio selection problem under\u0000generalized disappointment aversion (GDA). The implicit definition of the\u0000certainty equivalent within GDA preferences introduces time inconsistency to\u0000this problem. We provide the sufficient and necessary conditions for a strategy\u0000to be an equilibrium by a fully nonlinear ordinary differential equation (ODE).\u0000Through an exploration of the existence and uniqueness of solution to the ODE,\u0000we establish the existence and uniqueness of the equilibrium. Our findings\u0000indicate that under disappointment aversion (DA) preferences, non-participation\u0000in the stock market is the unique equilibrium. The numerical analysis reveals\u0000that, under GDA preferences, the investment proportion in the stock market\u0000consistently remains smaller than the investment proportion under the classical\u0000Expected Utility (EU) theory.","PeriodicalId":501045,"journal":{"name":"arXiv - QuantFin - Portfolio Management","volume":"56 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-01-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139481499","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Markowitz Portfolio Construction at Seventy","authors":"Stephen Boyd, Kasper Johansson, Ronald Kahn, Philipp Schiele, Thomas Schmelzer","doi":"arxiv-2401.05080","DOIUrl":"https://doi.org/arxiv-2401.05080","url":null,"abstract":"More than seventy years ago Harry Markowitz formulated portfolio construction\u0000as an optimization problem that trades off expected return and risk, defined as\u0000the standard deviation of the portfolio returns. Since then the method has been\u0000extended to include many practical constraints and objective terms, such as\u0000transaction cost or leverage limits. Despite several criticisms of Markowitz's\u0000method, for example its sensitivity to poor forecasts of the return statistics,\u0000it has become the dominant quantitative method for portfolio construction in\u0000practice. In this article we describe an extension of Markowitz's method that\u0000addresses many practical effects and gracefully handles the uncertainty\u0000inherent in return statistics forecasting. Like Markowitz's original\u0000formulation, the extension is also a convex optimization problem, which can be\u0000solved with high reliability and speed.","PeriodicalId":501045,"journal":{"name":"arXiv - QuantFin - Portfolio Management","volume":"44 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-01-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139423643","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Comparison of Markowitz Model and Single-Index Model on Portfolio Selection of Malaysian Stocks","authors":"Zhang Chern Lee, Wei Yun Tan, Hoong Khen Koo, Wilson Pang","doi":"arxiv-2401.05264","DOIUrl":"https://doi.org/arxiv-2401.05264","url":null,"abstract":"Our article is focused on the application of Markowitz Portfolio Theory and\u0000the Single Index Model on 10-year historical monthly return data for 10 stocks\u0000included in FTSE Bursa Malaysia KLCI, which is also our market index, as well\u0000as a risk-free asset which is the monthly fixed deposit rate. We will calculate\u0000the minimum variance portfolio and maximum Sharpe portfolio for both the\u0000Markowitz model and Single Index model subject to five different constraints,\u0000with the results presented in the form of tables and graphs such that\u0000comparisons between the different models and constraints can be made. We hope\u0000this article will help provide useful information for future investors who are\u0000interested in the Malaysian stock market and would like to construct an\u0000efficient investment portfolio. Keywords: Markowitz Portfolio Theory, Single\u0000Index Model, FTSE Bursa Malaysia KLCI, Efficient Portfolio","PeriodicalId":501045,"journal":{"name":"arXiv - QuantFin - Portfolio Management","volume":"11 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-01-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139423004","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}