A hybrid combinatorial approach to a two-stage stochastic portfolio optimization model with uncertain asset prices

Abstract

Portfolio optimization is one of the most important problems in the finance field. The traditional mean-variance model has its drawbacks since it fails to take the market uncertainty into account. In this work, we investigate a two-stage stochastic portfolio optimization model with a comprehensive set of real world trading constraints in order to capture the market uncertainties in terms of future asset prices. Scenarios are generated to capture uncertain prices of assets. Stability tests are performed and the results confirm the effectiveness of the scenario generation method used for this work. We propose a hybrid combinatorial approach, which integrates a hybrid algorithm and a linear programming (LP) solver for the problem with a large number of scenarios, where the hybrid algorithm is used to search for the assets selection heuristically and the LP solver solves the corresponding sub-problems of weight allocation optimally. The hybrid algorithm is based on Population Based Incremental Learning (PBIL) while local search, hash search, elitist selection, partially guided mutation and learning inheritance are also adopted. Comparison results against other 3 algorithms are given. The results show that our hybrid combinatorial approach can solve the two-stage stochastic model effectively and efficiently. The effects of different parameter settings are also examined.

Sustainable project portfolio selection under uncertainty with consider conflicting projects

“…The merits of this study utterly would be included as applying uncertain conditions, assessing various aspects of risk, and concentrating on economic criteria in the process of selection. ( Cui et al, 2020) suggested a two-stage stochastic portfolio optimization model with an extended set of real data trading constraints to tackle this issue regarding market uncertainty based on future asset price scenarios. In comparison to comprising models, it was concluded that the presented model is utterly reliable by employing the risk factors.…”

Model and efficient algorithm for the portfolio selection problem with real‐world constraints under value‐at‐risk measure

“…The well‐known mean‐variance model of Markowitz (1952) was one of the pioneering efforts in this field, and up to now, it has been extended from different aspects (Mencarelli and D'Ambrosio, 2018). One of the important extensions of the Markowitz model is the incorporation of real‐world constraints such as cardinality and position size constraints and transaction costs (Lwin et al., 2017; Cui et al., 2020; Juszczuk et al., 2023). As pointed out by Lwin et al.…”

“…In this approach, the future price of assets is described as a finite set of scenarios, and the decision‐making process is divided into two stages. In the first stage, before any uncertainty realization, the portfolio selection is decided, and in the second stage, the return of the constructed portfolio is realized under each scenario (Cui et al., 2020). As mentioned earlier, although creating a portfolio with the maximum expected return is important, decision‐makers usually concern about the investment risk.…”

Special issue on decision making and uncertainty

“…The paper by Zhang and Yang (2018) defines a new type of uncertain population growth model and uncertain logistic population growth model.The second category including seven papers applies uncertainty theory to solve various optimization problems. The two papers by andCui et al (2019) investigate different portfolio selection and optimization problems, respectively. The paper byChen and Zhu (2019) proposes an uncertain single period spatial oligopolistic electricity problem.…”