Article
Computer Science, Artificial Intelligence
Yuanzhen Dai, Zhongfeng Qin
Summary: This paper focuses on multi-period portfolio optimization in uncertain environments with minimum transaction lots, introducing a simplified additive format for total wealth, considering dynamic risk preferences of investors and building a mean-VaR model for risk control. Genetic algorithm is used to solve the model, with two numerical examples demonstrating the effectiveness of the approach.
APPLIED SOFT COMPUTING
(2021)
Article
Business, Finance
Yiying Liu, Yongbin Zhou, Juanjuan Niu
Summary: Sufficient description of stock returns is crucial for efficient portfolio optimization. Security returns are seen as random variables with sufficient historical data, and uncertain variables can enhance their effectiveness. This study focuses on optimizing transaction lots in uncertain trading environments, considering the changing risk preference of investors over the investment horizon. A genetic algorithm is used in an average-Value at Risk (VaR) framework to maximize wealth creation.
FINANCE RESEARCH LETTERS
(2023)
Article
Mathematics, Interdisciplinary Applications
Bo Li, Yayi Huang
Summary: Due to the complexity of security markets, there may be securities with massive effective data, invalid data, and insufficient data at the same time. This paper discusses the portfolio selection problem with different mental accounts under an uncertain random environment. It formulates an uncertain random model and presents two equivalent forms of the uncertain random portfolio model based on mental accounts. Numerical simulations are conducted to analyze the reality and practicability of the established models with two and three mental accounts.
CHAOS SOLITONS & FRACTALS
(2023)
Article
Computer Science, Artificial Intelligence
Guimei Tan, Xichang Yu
Summary: A new method - arc entropy, is proposed to measure the uncertainty of uncertain sets, and a new computational formula is introduced for quicker calculation while studying its properties and demonstrating its advantages in various applications.
JOURNAL OF INTELLIGENT & FUZZY SYSTEMS
(2022)
Article
Mathematics, Interdisciplinary Applications
Bo Li, Xiangfa Li, Kok Lay Teo, Peiyao Zheng
Summary: This paper proposes a new portfolio selection model involving uncertain and random return rates. By considering downside risks and diversification constraints, investment return and risk are quantified as uncertain random expected value and variance. The formulated model is transformed into equivalent deterministic models, and the NSGA-II algorithm is used to solve the bi-objective model, with a new optimal solution criterion proposed to find a single optimal solution in the Pareto optimal solution set.
CHAOS SOLITONS & FRACTALS
(2022)
Article
Mathematics
Lifeng Wang, Jinwu Gao, Hamed Ahmadzade, Zezhou Zou
Summary: The partial Gini coefficient is a measure of dispersion for uncertain random variables, serving as a risk measure and portfolio selection tool. In this paper, the authors define the partial Gini coefficient as a risk measure and provide a computational formula. They also apply it to characterize investment risk and study a mean-partial Gini model with uncertain random returns.
Article
Computer Science, Artificial Intelligence
Jin Liu, Jinsheng Xie, Hamed Ahmadzade, Mehran Farahikia
Summary: Entropy is a measure for quantifying uncertainty of random or uncertain variables in probability and uncertainty theory. The concept of exponential entropy is introduced to characterize uncertainty of uncertain variables and a formula is derived using inverse uncertainty distribution. Portfolio selection problems for uncertain returns are optimized using exponential entropy-mean models, with several examples provided for better understanding.
JOURNAL OF INTELLIGENT & FUZZY SYSTEMS
(2021)
Article
Computer Science, Artificial Intelligence
Yifan Zhang, Peilin Zhao, Qingyao Wu, Bin Li, Junzhou Huang, Mingkui Tan
Summary: Portfolio selection is a challenging task in finance, and this paper proposes a cost-sensitive method using deep reinforcement learning. The proposed method extracts price series patterns and asset correlations, and controls both transaction and risk costs effectively. Empirical results demonstrate its effectiveness and superiority in profitability, cost-sensitivity, and representation abilities.
IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING
(2022)
Article
Management
Sini Guo, Jia-Wen Gu, Christopher H. Fok, Wai-Ki Ching
Summary: This paper studies the online portfolio selection problem with transaction costs. It investigates the exact computation of transaction costs and derives upper and lower bounds to incorporate them into the optimal portfolio strategy. The paper also proposes the state-dependent exponential moving average method (SEMA) to accurately predict asset returns based on historical data and market states. Furthermore, it constructs the net profit maximization models (NPM and NPMRP) and combines them into the state-dependent online portfolio selection algorithm (SOPS). Empirical results show that the SOPS algorithm outperforms many state-of-the-art OLPS algorithms.
EUROPEAN JOURNAL OF OPERATIONAL RESEARCH
(2023)
Article
Green & Sustainable Science & Technology
Hongbo Li, Rui Chen, Xianchao Zhang
Summary: In this study, we propose a stochastic programming model for the uncertain public R&D project portfolio selection problem and transform it into an equivalent deterministic second-order cone programming model. Through simulation and computational experiments, we analyze the impacts of various factors on the project portfolio performance.
Article
Computer Science, Artificial Intelligence
Yufeng Li, Bing Zhou, Yingxue Tan
Summary: When investing in new stocks, predicting returns and risks without historical data is difficult. Therefore, a portfolio optimization model with an uncertain rate of return is proposed, using prospect theory and expected utility maximization. An improved GWO algorithm is designed to handle the complex characteristics of the model, outperforming traditional optimization algorithms.
COMPLEX & INTELLIGENT SYSTEMS
(2022)
Article
Computer Science, Artificial Intelligence
Guowei Jiang, Xiaoxia Huang, Tingting Yang
Summary: This paper examines the comparative static effects under uncertainty in a portfolio decision problem with both endogenous and background risks. By considering a situation where security return and background asset return cannot be reflected by historical data but are estimated by experts, an uncertain mean-chance model with background risk is proposed for optimal portfolio selection. The use of the chance of portfolio return failing to reach a threshold helps investors determine their risk tolerance and facilitates decision making. The solution to the programming problem under different threshold return levels and the effects of changes in mean and standard deviation of risky and background assets on investment decisions are analyzed. A real portfolio selection example is provided for illustration.
INTERNATIONAL JOURNAL OF INFORMATION TECHNOLOGY & DECISION MAKING
(2023)
Article
Computer Science, Artificial Intelligence
Zaiwu Gong, Xiaoxia Xu, Weiwei Guo, Enrique Herrera-Viedma, Francisco Javier Cabrerizo
Summary: This paper explores the uncertainty theory in group decision-making, proposing a theory that combines deterministic and indeterministic aspects in group decision-making. By using belief degrees and uncertainty distributions to accommodate individual preferences, different scenarios of uncertain minimum cost consensus models are discussed from various perspectives. Theoretical conditions for achieving consensus and analytical formulas for minimum total cost are deduced, with applications in carbon quota negotiation demonstrating the extension of existing models. In essence, the traditional models' conclusions can be seen as specific cases of the uncertain minimum cost consensus models under varying belief degrees.
INFORMATION FUSION
(2021)
Article
Computer Science, Artificial Intelligence
Weiwei Guo, Zaiwu Gong, Xiaoxia Xu, Ondrej Krejcar, Enrique Herrera-Viedma
Summary: This article explores the minimum-cost consensus model based on linear uncertainty distribution in uncertain theory, using uncertain distance measure and consensus utility to achieve flexible handling of group decision-making. The results show that uncertain consensus models are more adaptive and efficient in complex decision-making scenarios.
INFORMATION FUSION
(2021)
Article
Mathematics, Applied
Hong Seng Sim, Wendy Shin Yie Ling, Wah June Leong, Chuei Yee Chen
Summary: This paper proposes a sparse equity portfolio optimization model that aims to minimize transaction cost by avoiding small investments and promoting portfolio diversification. The model includes regularization of asset weights to promote sparsity, and a proximal method is used to handle the complexity of the objective function. An efficient algorithm is developed to find the optimal portfolio, and its global convergence is proven.
JOURNAL OF INEQUALITIES AND APPLICATIONS
(2023)