Stock Portfolio Selection Research Articles

Integration of prediction and optimization for smart stock portfolio selection

Machine learning (ML) algorithms pose significant challenges in predicting unknown parameters for optimization models in decision-making scenarios. Conventionally, prediction models are optimized independently in decision-making processes, whereas ML algorithms primarily focus on minimizing prediction errors, neglecting the role of decision-making in downstream optimization tasks. The pursuit of high prediction accuracy may not always align with the goal of reducing decision errors. The idea of reducing decision errors has been proposed to address this limitation. This paper introduces an optimization process that integrates predictive regression models within a mean–variance optimization setting. This innovative technique introduces a general loss function to capture decision errors. Consequently, the predictive model not only focuses on forecasting unknown optimization parameters but also emphasizes the predicted values that minimize decision errors. This approach prioritizes decision accuracy over the potential accuracy of unknown parameter prediction. In contrast to traditional ML approaches that minimize standard loss functions such as mean squared error, our proposed model seeks to minimize the objective value derived directly from the decision-making problem. Furthermore, this strategy is validated by developing an optimization-based regression tree model for predicting stock returns and reducing decision errors. Empirical evaluations of our framework reveal its superiority over conventional regression tree methods, demonstrating enhanced decision quality. The computational experiments are conducted on a stock market dataset to compare the effectiveness of the proposed framework with the conventional regression tree-based approach. Remarkably, the results confirm the strengths inherent in this holistic approach.

Two-stage stock portfolio optimization based on AI-powered price prediction and mean-CVaR models

With the advancement of prediction methods in the field of artificial intelligence, accurate price predictions can effectively support financial portfolio selection. This paper proposes an intelligent stock portfolio selection method based on a prediction neural network, incorporating signal processing and hyperparameter optimization techniques. The method is divided into two key stages: stock price prediction and portfolio selection. In the first stage, we apply Savitzky–Golay filtering to denoise price data and reveal its patterns, and optimize the hyperparameters of the long short-term memory network using the sparrow search algorithm to achieve high-precision stock price predictions. In the second stage, we use the mean-Conditional Value-at-Risk (mean-CVaR) model to select the optimal stock allocation, considering factors such as potential returns, prediction accuracy, and growth rate. Numerical comparisons based on multiple public financial datasets demonstrate that the proposed two-stage method significantly outperforms seven benchmark methods. Specifically, on the Shanghai and Shenzhen 300 (CSI 300) Index dataset, the proposed method achieves a determination coefficient of 0.9980 and an accuracy rate of 97.05%. Additionally, its cumulative returns reach 9.38%, 8.63%, and 7.54% at different confidence levels.

A novel regret-rejoice cross-efficiency approach for energy stock portfolio optimization

As climate change intensifies, more and more people realize that it is necessary and urgent to optimize energy structure through constructing energy asset portfolios. However, most of existing literature on energy asset portfolio models neglects real market restrictions, and the related portfolio performance evaluation approaches ignore the impact of the psychological factors of decision makers (DMs) on evaluation results, which limit their applications. To overcome these limitations, we first present a mean-WMCVaR (MWMC) model for energy stock portfolio selection with real features. Then, we employ the proposed MWMC model and its closely related works to yield a set of investment schemes and determine their characteristic index values including mean, WMCVaR, carbon intensity and ESG. Subsequently, we present a novel regret-rejoice cross-efficiency (NRRCE) approach to evaluate the performance of these investment schemes, and then make a comparison with the ones of the other seven competitive evaluation approaches by using an empirical example. Empirical results show that the proposed MWMC model can yield an optimal investment scheme with a better trade-off between economic and ecological benefits, and our NRRCE approach can effectively provide reasonable and reliable evaluation results for different investment schemes.

The Implementation of Roy's Safety-First Criterion in Stock Portfolio Selection

Statistical data shows that the Indonesian capital market has experienced significant growth. This growth is attributed to public awareness of the benefits of stock investments. However, with an increasing number of new investors entering the stock market, attention to investment risks deepens. Many investors prefer stocks that are easily predictable and have low risk, as higher volatility increases the level of uncertainty in obtaining returns. The relatively high risk level in investing requires investors to minimize risks, one of which is by diversifying funds into various investment assets, commonly known as optimal portfolio selection. An optimal portfolio can be formed using various methods and approaches. One method for portfolio selection is the application of the Safety First Criterion, a method dependent on downside risk, referring to risks that result in losses. This article conducts a simulation of the implementation of Roy's Safety First Criterion using stock data from the largest companies in eleven sectors over the past year. These sectors include basic materials, communication services, consumer cyclical, consumer defensive, energy, financial services, healthcare, real estate, technology, and utilities. Based on this analysis, out of the eleven stocks, six stocks meet Roy's criteria: LIN, GOOG, AMZN, BRK-B, LLY, and AAPL, with respective weights of LIN=10.84%, GOOG=12.61%, AMZN=24.67%, BRK-B=1.37%, LLY=34.17%, and AAPL=16.35%. Using Roy's Safety First Criterion for selected stocks from various sectors indicates that the resulting portfolio has a very low risk level, specifically 1%. This means that by using the portfolio obtained from the eleven stocks, investors can achieve minimal risk, making this portfolio secure for new investors.

Spatial-Temporal Stock Movement Prediction and Portfolio Selection based on the Semantic Company Relationship Graph

Spatial-Temporal Stock Movement Prediction and Portfolio Selection based on the Semantic Company Relationship Graph

A novel randomized weighted fuzzy AHP by using modified normalization with the TOPSIS for optimal stock portfolio selection model integrated with an effective sensitive analysis

Building an ideal stock portfolio is a difficult and important undertaking in the world of financial decision-making. The Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS) technique and the Randomised Weighted Fuzzy Analytic Hierarchy Process (AHP) are combined in this research to create a powerful model for selecting the best stocks for a portfolio. To improve the portfolio's robustness under various market scenarios, our model also incorporates sensitivity analysis. Given the inherent uncertainty in financial markets, the suggested approach starts by using modified fuzzy AHP based on introducing the randomized weights and the novel normalization to get expert views and assign suitable weights to the underlying selection criteria. Based on the established criteria and associated fuzzy scores, the TOPSIS approach is then used to rank and choose the best stocks. Our model also includes sensitivity analysis to evaluate the performance of the portfolio under various market conditions. Extensive empirical tests are performed utilizing historical stock market data to assess the effectiveness of our strategy. The outcomes show the suggested model's superiority in building stock portfolios that beat conventional approaches in terms of risk-adjusted returns.

Deep learning in stock portfolio selection and predictions

Deep learning (DL) has made its way into many disciplines ranging from health care to self-driving cars. In financial markets, we see a rich literature for DL applications. Particularly, investors require robust algorithms that can navigate and make sense of extremely noisy and volatile markets. In this work, we use deep learning to select a portfolio of stocks and use a genetic algorithm to optimize the hyperparameters of DL. The work analyzes the improvement in using genetic-based hyperparameter optimization over grid searches. The Genetic Algorithm brings 40% improvements in prediction when compared to a random-grid search. Novelty-wise, the work couples a genetic-based hyperparameter optimization with multiple Deep RankNet models to predict the behavior of financial assets. Our results show promising portfolio returns 20% better than the general market. In the highly volatile COVID 19 period, the models exceed market returns by more than double. Overall, this paper brings a comprehensive work that integrates hyperparameter optimization, Deep RankNet, LSTM, period size variations, input variable transformation, feature selection, training/evaluation ratio analysis, and multiple portfolio selection strategies.

A time-varying stock portfolio selection model based on optimized PSO-BiLSTM and multi-objective mathematical programming under budget constraints

A time-varying stock portfolio selection model based on optimized PSO-BiLSTM and multi-objective mathematical programming under budget constraints

A Hybrid Model Based on Multi-Attribute Decision Making and Fuzzy Multi-Objective Mathematical Programming for Stock Portfolio Evaluation and Selection Using Z-Number Approach

A Hybrid Model Based on Multi-Attribute Decision Making and Fuzzy Multi-Objective Mathematical Programming for Stock Portfolio Evaluation and Selection Using Z-Number Approach

Optimal Selection of Stock Portfolios Using Multi-Criteria Decision-Making Methods

In the past, investors used their own or others’ experiences to achieve their goals. With the development of financial management, investors’ choices became more scientific. They could select the optimal choice by using different models and combining the results with their experiences. In portfolio optimization, the main issue is the optimal selection of the assets and securities that can be provided with a certain amount of capital. In the present study, the problem of optimization, i.e., maximizing stock portfolio returns and minimizing risk, has been studied. Therefore, this study discussed comprehensive modeling for the optimal selection of stock portfolios using multi-criteria decision-making methods in companies listed on the Tehran Stock Exchange. A sample of 79 companies listed on the Tehran Stock Exchange was used to conduct this research. After simulating the data and programming them with MATLAB software, the cumulative data analysis model was performed, and 24 companies were selected. This research data were collected from the financial statements of companies listed on the Tehran Stock Exchange in 2020. The primary purpose of this study was a comprehensive modeling for the optimal selection of stock portfolios using multi-criteria decision-making methods in companies listed on the Tehran Stock Exchange. The index in the Tehran Stock Exchange can be used to provide a comprehensive and optimal model for the stock portfolio; different multi-index decision-making methods (TOPSIS method), the taxonomy method (Taxonomy), ARAS method, VIKOR method, The COPRAS method and the WASPAS method can all identify the optimal stock portfolio and the best stock portfolio for the highest return.

A novel hybrid simplified group BWM and multi-criteria sorting approach for stock portfolio selection

Many real-life issues can be modeled as multi-criteria sorting problems. In this type of problem, a set of options are assigned to predefined classes. This research aims to propose a new hybrid approach including simplified group best–worst and multi-criteria sorting methods to classify options considering decision-maker constraints. It is crucial to consider such constraints, and ignoring them can make the results ineffective for the decision-maker. The method developed in this study for weighting the criteria can be considered the most unmixed group best–worst method ever proposed, as no optimization model and specialized software are needed to solve it. In this research, a multi-criteria sorting method is also suggested. The developed method compares alternatives with limiting profiles and then classifies alternatives based on their distance from the limiting profiles. This method can consider investor preferences, such as the number of alternatives in categories or the maximum number of stocks allowed from each industry in the stock portfolio. The proposed approach is used in a real case study of stock portfolio selection in the Tehran Stock Exchange. The results of the methods proposed in this study were compared with the results of the previous methods using numerical examples and a real case study. The results show that the developed methods are valid and accurate. The suggested approach helps managers to include their constraints in the decision-making process and can be applied to real-world problems with a classification nature.

Optimal Stock Portfolio Selection with a Multivariate Hidden Markov Model

Optimal Stock Portfolio Selection with a Multivariate Hidden Markov Model

Development of Optimal Stock Portfolio Selection Model in the Tehran Stock Exchange by Employing Markowitz Mean-Semivariance Model

In an increasingly complex financial market, selecting the optimal stock portfolio has become a subject of intense debate. This study aims to develop a model for optimal stock portfolio selection. We apply Markowitz's mean-semivariance approach to determine the downside risk of portfolios, which reflects investors' intuitive perception of risk. In the first stage, the combination of the Analytic Hierarchy Process (AHP) and Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS) with interval data is employed to identify and rank good quality stocks according to the recommended criteria by experts. After selecting qualified stocks, in the second phase, we create portfolios, and the weight invested in each stock is determined. Then, three portfolios are created for three groups of risk-averse, neutral to risk, and risk-taker investors. The mean-semivariance optimization model is used in this phase. The proposed approach in the paper is implemented in a real case study of the Tehran stock exchange (TSE). Three portfolios for three groups of investors were evaluated and compared to the market performance using sharp criteria. All three portfolios outperformed the market portfolio both in terms of risk and return. The proposed model of this study can be utilized as a decision support tool when forming an optimal stock portfolio by considering both experts’ opinions on stock evaluation and investor risk preferences simultaneously.

Risk Reduction by Stock Portfolio Selection using LSTM and K-means Clustering

The stock market is very complex to understand. In the stock market, the stock prices keep fluctuating and it is tough to track the price of stocks. Most users find it difficult to choose the right stocks for their investment and there is a lot of confusion when selecting the right stocks in fear of losing your hard-earned money. In this paper, we discuss how to create a stock portfolio on the basis of stock prediction and then select multiple stocks using K-means clustering to reduce risk as per the profit. Many people aren’t aware of price fluctuation and lose their money in specific stocks. This paper helps to understand the benefit of diversification of stock portfolios to reduce the risk of stock market fluctuations.

Stock Selection Modeling and Portfolio Selection in Emerging Markets

This article addresses stock selection modeling and portfolio selection and implementation in EMs. The authors view EM investing as a special case of global investing, demonstrating how stock-selection models in EMs with price momentum and forecast earnings acceleration factors can enhance returns. The authors construct index-enhanced portfolios for EMs that offer superior return-to-risk ratios relative to domestic portfolios. On the basis of back test and in real-time performance, the authors show how an EM portfolio using forecast earnings and price momentum anomalies can be developed, estimated, and implemented to generate statistically significant excess returns.

The Performance of Shrinkage Estimator for Stock Portfolio Selection in Case of High Dimensionality

Harry Markowitz introduced the Modern Portfolio Theory (MPT) for the first time in 1952 which has been applied widely for optimal portfolio selection until now. However, the theory still has some limitations that come from the instability of covariance matrix input. This leads the selected portfolio from MPT model to change the status continuously and to suffer the high cost of transaction. The traditional estimator of the covariance matrix has not solved this limitation yet, especially when the dimensionality of the portfolio soars. Therefore, in this paper, we conduct a practical discussion on the feasible application of the shrinkage estimator of the covariance matrix, which is expected to encourage the investors focusing on the shrinkage–based framework for their portfolio selection. The empirical study on the Vietnam stock market in the period of 2011–2021 shows that the shrinkage approach has much better performance than other traditional methods on the primary portfolio evaluation criteria such as return, level of risk, Sharpe ratio, maximum loss, and Alpla coefficient, especially the superiority is even more evident when the dimension of covariance matrix increases. The shrinkage approach tends to create more stable and secure portfolios than other estimators, as demonstrated by the average volatility and maximum loss criteria with the lowest values. Meanwhile, the factor model approach is able to generate portfolios with higher average returns and lower portfolio turnover; and the traditional approach gives good results in the case of low—dimensionality. Besides, the shrinkage method also shows effectiveness when beating the tough market benchmarks such as VN-Index and 1/N portfolio strategy on almost performance metrics in all scenarios.

Stock portfolio selection balancing variance and tail risk via stock vector representation acquired from price data and texts

Recent works on portfolio selection report ways to incorporate textual data in addition to price movements. Price, texts, and events as what lies underneath take heterogeneous data form and therefore have been processed without any consistent mathematical formulation.In this article, we propose to generalize portfolio selection by representing all related objects (stocks, news, events) in an embedding vector space, that we call a NEws-STock space with Event Distribution (NESTED). A NESTED forms an inner product vector space (Hilbert space), in which texts and stocks are represented as vectors (embeddings), acquired through a distribution of events. In this article, we first theoretically reformulate Markowitz’s portfolio optimization problem on NESTED. We show how our new formulation has the potential to better incorporate the tail risk, which is represented better in textual data.One typical method to acquire such embeddings is via neural computing. Our experimental results, obtained by using it on 24 news-price datasets across three markets, showed that the Pareto’s exponent in the negative tail of the generated portfolios increased in all markets, which is evidence that the stock embeddings captured the tail risks. Our method showed a large improvement balancing between the tail risk and non-tail risk, up to 45.5% larger gain and 59.4% larger Information ratio.

Stock portfolio selection using a new decision-making approach based on the integration of fuzzy CoCoSo with Heronian mean operator

The main objective of stock portfolio selection is to distribute capital to selected stocks to get the most profitable returns at a lower risk. The performance of a stock depends on a number of criteria based on the risk-return measures. Therefore, the selection of shares is subject to fulfilling a number of criteria. In this paper, we have adopted an integrated approach based on the two-stage framework. First, the heronian mean operator (improved generalized weighted heronian mean and improved generalized geometric weighted heronian mean) is combined with the traditional Combined compromise solution (CoCoSo) method to present a new decision-making model for dealing with stock selection problem. Second, Base-criterion method is used to calculate the relative optimal weights of the specified decision criteria. Despite the uncertainties, the advanced CoCoSo-H model eliminates the efficacy of anomalous data and make complex-decisions more flexible. A case study of stock selection for portfolio under National stock exchange (NSE) is discussed to validate the applicability of the proposed model. Different portfolio () have been constructed using Particle swarm optimization (PSO). The outcome shows the prominence and stability of the proposed model when compare to previous studies.

Using ELECTRE-TRI and FlowSort methods in a stock portfolio selection context

In recent years, multi-criteria sorting problems have become an interesting topic for researchers working on multi-criteria decision-making. ELimination and Choice Expressing REality (ELECTRE)-TRI and FlowSort are well-known approaches suggested for such a classification. The current study aimed to implement ELECTRE-TRI and FlowSort methods in the stock portfolio selection (SPS) as one of the most popular and important decision-making subjects and compare the outcomes of each method to understand how these methods perform in SPS problems. In this study, the best–worst method was applied to determine the weights of criteria. Four approaches for ELECTRE-TRI and 15 approaches for FlowSort were considered. Finally, 19 different approaches were considered to select stocks from a large pool of stocks. Results indicated that the model parameter should be properly defined to minimize inconsistencies and improve the power of the model.

Deep Learning in Stock Portfolio Selection and Predictions

Deep Learning in Stock Portfolio Selection and Predictions