Stock Portfolio Optimization Research Articles

A hybrid approach for portfolio construction: Combing two‐stage ensemble forecasting model with portfolio optimization

AbstractCombining the stock prediction with portfolio optimization can improve the performance of the portfolio construction. In this article, we propose a novel portfolio construction approach by utilizing a two‐stage ensemble model to forecast stock prices and combining the forecasting results with the portfolio optimization. To be specific, there are two phases in the approach: stock prediction and portfolio optimization. The stock prediction has two stages. In the first stage, three neural networks, that is, multilayer perceptron (MLP), gated recurrent unit (GRU), and long short‐term memory (LSTM) are used to integrate the forecasting results of four individual models, that is, LSTM, GRU, deep multilayer perceptron (DMLP), and random forest (RF). In the second stage, the time‐varying weight ordinary least square model (OLS) is utilized to combine the first‐stage forecasting results to obtain the ultimate forecasting results, and then the stocks having a better potential return on investment are chosen. In the portfolio optimization, a diversified mean‐variance with forecasting model named DMVF is proposed, in which an average predictive error term is considered to obtain excess returns, and a 2‐norm cost function is introduced to diversify the portfolio. Using the historical data from the Shanghai stock exchange as the study sample, the results of the experiments indicate the DMVF model with two‐stage ensemble prediction outperforms benchmarks in terms of return and return‐risk characteristics.

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.

Quantitative stock portfolio optimization by multi-task learning risk and return

Selecting profitable stocks for investments is a challenging task. Recent research has made significant progress on stock ranking prediction to select top-ranked stocks for portfolio optimization. However, the stocks are only ranked by predicted stock return, ignoring the stock price volatility risk — a critical aspect for stock selection and investments. Moreover, they preliminarily attempted to capture the effects of related stocks from a singular relation, disregarding the rich information regarding multiple spillover effects from related stocks and the distinctions in effects among various relations. Thus, we propose a risk and return multi-task learning model with a heterogeneous graph attention network (HGA-MT) to predict stock ranking for portfolio optimization. First, to aggregate the multiple spillover effects of related stocks, we introduce graph convolutional networks to fuse the effects of related stocks in each relation and design an attention network to allocate varying weights to different types of relationships. Second, we use a multi-task learning paradigm to learn stock return and volatility risks jointly. The stock ranking results are calculated by simultaneously considering the risk and return. Thus, Top-K ranked stocks are recommended in the portfolio for the next trading day to achieve higher and more stable profits. Extensive experiments prove that HGA-MT outperforms previous state-of-the-art methods in stock ranking and backtesting trading evaluation tasks.

Mean-Value-at-Risk Portfolio Optimization Based on Risk Tolerance Preferences and Asymmetric Volatility

Investors generally aim to obtain a high return from their stock portfolio. However, investors must realize that a high value-at-risk (VaR) is essential to calculate for this aim. One of the objects in the VaR calculation is the asymmetric return volatility of stocks, which causes an unbalanced decrease and increase in returns. Therefore, this study proposes a mean-value-at-risk (mean-VaR) stock portfolio optimization model based on stocks’ asymmetric return volatility and investors’ risk aversion preferences. The first stage is the determination of the mean of all stocks in the portfolio conducted using the autoregressive moving average Glosten–Jagannathan–Runkle generalized autoregressive conditional heteroscedasticity (ARMA-GJR-GARCH) models. Then, the second stage is weighting the capital of each stock based on the mean-VaR model with the investors’ risk aversion preferences. This is conducted using the Lagrange multiplier method. Then, the model is applied to stock data in Indonesia’s capital market. This application also analyzed the sensitivity between the mean, VaR, both ratios, and risk aversion. This research can be used for investors in the design and weighting of capital in a stock portfolio to ensure its asymmetrical effect is as small as possible.

Decision Support Systems in Stock Investment Problems

This study compiles decision support systems that aim to optimize financial decision processes by examining the literature studies targeting stock investments. The review encompasses a range of methodologies and applications, from traditional approaches such as Markowitz’s Modern Portfolio Theory, Black-Litterman, and Single Index models to artificial intelligence-based techniques. In detail, the contributions of Decision Support Systems to stock portfolio construction and portfolio optimization processes along with comparative analyses between these systems are scrutinized. The review also aims to enable researchers and practitioners to be engaged in portfolio optimization with a framework for future investigations in areas such as historical data analysis, future price movement prediction, assessment of risk factors, and determination of optimal portfolio distribution. Furthermore, it seeks to enhance the understanding of decision support systems employed in portfolio optimization, facilitating a more comprehensive grasp of their utility within stock investments.

Can Lexicographic Goal Programming, Artificial Neural Networks, and Value-at-risk Methods be Effective in Analysis of 50 Highest Trade Frequency Issuers Optimum Portfolio

This study aims to describe the optimal stock portfolio as a series of processes for selecting a stock portfolio that suits the needs and desires of investors. The variable used is the daily closing price of the stock exchanges listed on the Indonesia Stock Exchange for the period September 1, 2016 to September 1, 2022 for 50 stocks with the highest trading frequency in the first quarter of 2022 using the value at risk, lexicographic goal method, programming, neural networks, and artificial neural networks. It is known that the value-at-risk method in generating shortfalls is carried out using a valuable approach that exceeds the threshold where the assumption of a 5% tail minimum return is generalized with the Pareto distribution. The input of the neural network in producing the output obtained through the PACF and ACF plots is t1. Issuer price forecasts are obtained using the neural network method and the backpropagation algorithm for each issuer, an SSMS of Rp. 1808.3, ICBP of Rp. 12,717, ASII Rp. 6832.1, BBNI of Rp. 6234.6, INCO of Rp. 2781.8, CPIN of Rp. 12,763, TLKM worth 2630.1, SCMA Rp. 2885.5, and WSKT worth Rp. 1575.7. The issuers obtained have a small MAPE value, which is below 10%. Programming the lexicographical objectives resulted in the nine selected issuers having a minimum standard of risk with maximum returns. Of the nine stocks selected, CPIN, SSMS, BBNI, ICBP, TLKM, ASII, INCO, SCMA and WSKT.
 Keywords: return, value at risk, lexicographic goal programming, backpropagation

Optimal Stock Portfolio Establishment with Active and Passive Strategy Using Price to Book Value and Price Earning to Growth Ratio Approach in IDX30 Index 2013-2018 Period

The volatility phenomenon of stock returns shows the return and risk faced by investors in investment activities. One method that can be used by investors to get maximum profit while compressing the risk into the minimum level is by diversifying its investments through portfolios. This study aims to determine the simulation results of optimal stock portfolio establishment with active and passive strategy using Price to Book Value and Price Earning to Growth ratio approach and the results of the Sharpe, Treynor, and Jensen performance evaluations on the established portfolio. This research appertains in quantitative research. The object of the research was the IDX30 Index and 17 of 30 companies which consistently classified into the IDX30 Index for the 2013-2018 period were selected as the research sample. The results showed that High PEG consistently provides better than the average returns and risks, whether in passive strategy, annual’s active strategy, or semester’s active strategy. Whereas, High PBV in the passive and annual's active strategy showed a high rate of return above the average, while in the semester active strategy, showed the lowest level of risk. Overall the semester's active strategy has the highest accumulated rate of return with a relatively low rate of risk. This result match with the purpose of optimal portfolio establishment. Moreover, the results of the performance evaluation showed that in the semester's active strategy, High PEG gives the best score based on the results of the performance evaluation of Sharpe, Treynor and Jensen index.

Stock Analysis and Portfolio Optimization

This paper investigates financial time series from US stock markets from a quantitative perspective. The returns of all stocks are clustered with K-means with five centroids; in each group, the store with the maximum return is selected. For all five chosen supplies, construct the portfolio with different stock weights and optimize the combination with Monte Carlo to mitigate risk and maximize the Sharpe ratio. Comparing the weights under different stock weights concludes that the optimal portfolio can be obtained with the maximized Sharpe ratio scenario. More interestingly, the consequences based on market value also give remarkable cumulative returns.

Stock Portfolio Optimization Using Mean-Variance and Mean Absolute Deviation Model Based On K-Medoids Clustering by Dynamic Time Warping

The tendency of investors to choose investments with maximum return and minimal risk causes the need for diversification in a portfolio to form an optimal portfolio. A lot of research on stock portfolio optimization has been conducted extensively, but not many have tried to apply machine learning concepts such as clustering analysis to accelerate the establishment of a model that can have a positive effect on the time and cost efficiency of portfolio management. However, clustering is only limited to determining the optimal stock candidate, so it is necessary to add another optimization model to calculate the portfolio weight. Based on these problems, this study carried out portfolio optimization using Mean-Variance (MV) and Mean Absolute Deviation (MAD) model based on K-Medoids Clustering by Dynamic Time Warping approach using Monte Carlo-Expected Tail Loss for risk analysis. Based on the analysis results, the MAD portfolio is more optimal than the MV portfolio by the MAD portfolio consists of five stocks, namely BMRI shares with a weight of 0.06243, UNTR shares of 0.08658, BBRI shares of 0.10285, BBCA of 0.53623, and KLBF shares of 0.21191 are the best optimal portfolios. The optimal portfolio of the MAD model has a rate of return of 87.836% in May 2017 - December 2022 with a portfolio performance of 0.03704, while the resulting risk level based on Carlo-Expected Tail Loss is 2.2416%.

Implementation of Lo Kheng Hong Investment Strategy to Build Optimal Stock Portfolio

The favorite investment choice of the Indonesian people, gold is in first place with a survey of 60.2%, and stock investment is ranked sixth with a value of 22.5%. Compared to Indonesia’s total population, the number of individual investors has a relatively small percentage of 1.48% but continues to show significant growth yearly. Before entering the Indonesia Stock Exchange, every stock investor must have good knowledge and strategies to get optimal returns and risks. Lo Kheng Hong is an individual investor who has proven successful in the Indonesian stock exchange. The purpose of this research is to get an overview of the results of stock investment using the Lo Kheng Hong investment strategy on the Indonesian stock exchange. The method used to select stocks IDX is to use the PBV <0.5, PER <10, and Market Cap > 500 billion; after that, the top 4 stocks with the largest Total Annual Income were selected in 2012. The author uses the RTI Analysis Method (Read Think Analysis), which Lo Kheng Hong uses to verify that the stocks being screened are truly undervalued stocks with good fundamentals. Optimize portfolio and investment risk using the Excel Solver tool to obtain the composition of the stock weight with the best Sharpe Ratio. The author simulated purchasing shares within ten years using the buy hold, rebalancing method of buying shares every 1, 3, 6, and 12 months. Then, compare the Portfolio performance with the Composite Stock Price Index (IHSG). The results obtained from this study are the composition of the PTRO stock portfolio = 15.01%, INKP = 44.19%, SMDR = 35.90%, and TKIM = 5% by maximizing the Sharpe ratio of 0.2134. For the method of buying shares using the rebalancing method every year, it gets the best performance compared to the buy and holds or rebalancing plans for other periods, with a yield of 3400.61% and a Sharpe ratio of 0.1807. Portfolio performance using the Lo Kheng Hong investment strategy can exceed the Jakarta Composite Index (IHSG) performance within a 10-year investment period.

Application of Lexicographic Goal Programming Method on Stock Portfolio Optimization With Expected Shortfall Approach

The purpose of this study is to form several optimal portfolios based on the proportion of funds invested in stocks, then the most optimal portfolio will be selected among the optimal portfolios that have been formed. The method used in this study is the lexicographic goal programming method, which is to determine the optimal portfolio based on the proportion of invested funds, then the selection of the most optimal portfolio is determined using the expected shortfall method. This study used data from 5 companies contained in Indonesia Stock Exchange (IDX). The results showed that from the proportion of invested funds, 11 optimal portfolios were obtained. Using the expected shortfall method, the most optimal portfolio of 11 portfolios was obtained, namely type 5 portfolio.

An Accelerated Optimization Approach for Finding Diversified Industrial Group Stock Portfolios with Natural Group Detection

Stock portfolio optimization is always an interesting and attractive research problem due to the variety of stock markets. To find a useful stock portfolio, metaheuristic-based approaches have been presented to obtain diverse group stock portfolios (DGSPs) by considering the diversity of stock portfolios in the past. However, in the existing DGSP algorithms, two problems remain to be solved. The first is how to set a suitable group size, and the second is that the evolution process is time-consuming. To solve these problems, in this paper, an approach using grouping genetic algorithms (GGAs) was proposed for optimizing a DGSP. For setting a suitable group size, the proposed approach utilized two attributes of group stocks, including the return on equity and the price/earnings ratio. Then, to derive better stock groups, a cluster validation factor was designed, which was used as part of a fitness function. To solve the time-consumption problem, using the designed temporary chromosome, the number of stock portfolios that need to be evaluated could be reduced in the proposed approach to speed up the evolution process. Finally, experiments on two real stock datasets containing 31 and 50 stocks were conducted to show that the proposed approach was effective and efficient. The results indicated that the proposed approach could not only achieve similar returns but also accelerate the evolution process when compared with the existing algorithms.

Portfolio Optimization Based on Return Prediction and Semi Absolute Deviation (SAD)

A portfolio is a collection of investment financial assets managed by financial institutions or individuals. In investment activities, investors expect minimal loss risk and optimal stock portfolio weight to get maximum profit. Investors can monitor changes in stock index values to compare portfolio performance. This research has discussed how to build a portfolio based on stock datasets with the LQ45 index using return predictions from the artificial neural network (ANN) method with semi-absolute deviation (SAD). Furthermore, the portfolio is optimized by looking for weights that match it. After that, a comparison of portfolio performance was carried out using the Sharpe ratio (SR) method between the semi-absolute deviation (SAD) portfolio and the portfolio resulting from the formation of the equal weight (EW) portfolio. Portfolio performance with ANN prediction and SAD is better than equal-weight portfolios in terms of mean return, standard deviation, and sharpe ratio for portfolios with few stocks, namely 2 and 3 stocks. In addition, a portfolio with a higher number of stocks can make the portfolio value from the ANN close prediction algorithm process and the selection of weights based on SAD is better than portfolios with equal weight for each list of stocks in the portfolio.

Is the Markowitz Model Still Effective in Forming an Optimal Stock Portfolio? Empirical Evidence from the IDX ESG Leaders Index

The current portfolio calculation was basically developed by Markowitz (1952) who gave rise to the Modern Portfolio Theory. However, along with the development of science, other models emerged in portfolio calculation. This study tried to provide a flashback to whether the Markowitz portfolio model is still effective in forming an optimal portfolio compared to other portfolio calculation models. This study used descriptive research with a quantitative approach. This study used secondary data in the form of daily stock prices of companies, including the IDX ESG Leaders Index on the Indonesia Stock Exchange for the period of December 2020 to March 2022. The results showed that 11 stocks out of 25 stock samples were included in the optimal portfolio in both the Markowitz model and the Single Index Model but with different stock compositions. In terms of portfolio performance, the Markowitz model is still superior to the Single Index Model. This can be seen from the calculation results of the three indexes used, including the Sharpe Index, Treynor Index, and Jensen Index. This study provided an important contribution that the Markowitz model portfolio formation can be used and has proven to perform well in the formation of a stock portfolio.

Analisis Kinerja Portofolio Saham Menggunakan Metode Sharpe, Treynor dan Jensen

The purpose of this study was to determine the optimal stock portfolio performance on the LQ45 stock index on the Indonesia Stock Exchange for the period February 2019- December 2022. The data analysis method uses a single index model to form an optimal portfolio and uses the Sharpe, Treynor, and Jensen index methods to assess performance stock portfolio. Based on the results of the analysis of optimal portfolio formation, it shows that out of 30 samples of LQ45 index stocks, 8 stocks were obtained which included optimal portfolio stocks, namely ERB values > C*, namely ADRO, ANTM, BBCA, BRPT, INCO, INDY, ITMG, & MEDC stocks. the results of the stock portfolio performance appraisal analysis show that of the three portfolio performance appraisal index methods, the calculation of stock performance analysis, the average value of the Sharpe method is 0.1241, while the treynor method is 0.0112, and the Jensen method is 0.0334. The highest performance of the third method is the Sharpe method with an average performance value of 0.1241. So the Sharpe method is the best performance method compared to the Treynor and Jensen methods. The higher the Sharpe index value, the better the performance of the stock portfolio.

STUDI KOMPARASI PEMBENTUKAN PORTOFOLIO OPTIMAL MENGGUNAKAN MODEL MARKOWITZ DAN MODEL INDEKS TUNGGAL

The purpose of this study was to determine the results of the comparison of the formation of the optimal portfolio using the Markowitz model and the Single Index model. This study was conducted on companies that are members of the LQ45 Index in the period August 2017 to January 2020. The number of companies used as samples were 32 companies using a purposive sampling method, namely data selection based on several criteria. The data were obtained by observing, understanding, and studying books, journals, theses and other literature related to research, as well as from the closing price of each company's shares at Yahoo Finance and idx.co.id. The data analysis technique used is the Wilcoxon-Mann-Whitney test. The results of this study showed that is no significant difference in the average return of the two models, so it can be concluded that to form an optimal LQ45 stock portfolio, both methods can be used, namely the Markowitz model and the Single Index model. The result of study have implications for the selection of stocks for investors. Investors can use the Markowitz model and the Single Index model in selecting the optimal portfolio because the two methods are the same but only differ in the calculation.

Analysis of Stock Portfolio Optimization in the Telecommunications Sector Using the Single Index Model

The purpose of this study was to determine the optimal portfolio in the telecommunications sector listed on the Indonesia Stock Exchange based on the Jakarta Composite Index for the period January 2018–December 2020 using the Single Index Model. This type of research is an applied research. This type of research is applied research with secondary data obtained from the Indonesia Stock Exchange, Yahoo Finance, and Bank Indonesia. The number of samples taken is 5 stocks, namely TLKM, ISAT, EXCL, BTEL, and FREN. Based on the results of the analysis of the 5 stocks that are members of the JCI, the combination of 2 stocks that make up the optimal portfolio, namely ISAT and FREN, produces an expected return of 5.08% with a risk of 8.02%.

Pembentukan Portofolio Optimal Saham berdasarkan Model Markowitz dan Model Indeks Tunggal pada Perusahaan IDX BUMN20 Tahun 2017-2021

This study aims to analyze the formation of the optimal stock portfolio based on the Markowitz model and the Single Index model in the IDX BUMN20 company in 2017- 2021. The type of research used is applied quantitative descriptive. The secondary data used are the monthly closing price of shares and the monthly JCI as well as the interest rate of Bank Indonesia. The sample in this study were 15 companies. The model used is the Markowitz model and the Single Index model. The results of the analysis show that 1. There are 11 stocks that are included in the optimal portfolio candidates, namely ANTM, BBTN, BBNI, BBRI, BMRI, JSMR, PGAS, PTBA, SMGR, TINS, and TLKM 2. The formation of the optimal portfolio of the Markowitz model consists of eight issuers with the proportions are BMRI 20%, TLKM 20%, BBRI 18%, PTBA 15%, ANTM 11%, JSMR 9%, SMGR 4% and BBTN 3% by providing a portfolio risk level of 0.0002 or 0.02% and the expected return on the portfolio is 0.0050 or 0.5%. 3. The formation of the optimal portfolio of the Single Index model consists of three issuers with the proportions of BBRI at 48%, ANTM at 45%, TINS 7%.

Моделювання оптимального інвестиційного портфеля орієнтованого на мінімізацію ризику

Abstract. Introduction. Portfolio optimization modelling is a study aimed at determining the best way to allocate an investor's funds between securities of different companies. The study uses a number of methods and approaches, such as: multicriteria optimization methods - optimization of criteria for minimum and maximum, modeling of the optimal stock portfolio with minimal risk, to determine the best way to allocate funds that will help investors achieve maximum return while minimizing risk. Purpose. There are a lot of publicly traded companies in the modern American stock market. Having chosen the companies that may grow in the future, the problem of allocating your funds among them remains relevant. Building an optimal portfolio of securities using various economic and mathematical methods solves the problem of allocating financial resources, focusing on the desired future profit and the level of risk exposure. Results. The model for this study was built using one of the methods of multi-criteria optimization - criteria convolution, taking into account portfolio diversification and the specified constraints. The optimization is based on the "Modern Portfolio Theory" of the prominent scientist Harry Markowitz. Conclusions. As a result, we built an optimal and diversified portfolio of shares, in which each company on the list represents at least 1%. All constraints have been met and the main conditions have been fulfilled - the portfolio minimizes risk and maximizes profit. With a minimum risk of 5.39%, we expect a return of 1.75%. Such results can be obtained if we use a convolution of the criteria where the preference is given to minimizing risk - 0.7. The largest contributors to the portfolio were the following companies: T-Mobile Us Inc., McKesson Corporation, The Kroger Co., Microsoft Corporation, and Apple Inc. These companies account for a significant portion of the portfolio - 65%. Given the focus on portfolio diversification and the ratio of risk to potential return, we can say that the portfolio is efficient and can be used in practice.

Forecasting Stock Market Indices Using the Recurrent Neural Network Based Hybrid Models: CNN-LSTM, GRU-CNN, and Ensemble Models

Various deep learning techniques have recently been developed in many fields due to the rapid advancement of technology and computing power. These techniques have been widely applied in finance for stock market prediction, portfolio optimization, risk management, and trading strategies. Forecasting stock indices with noisy data is a complex and challenging task, but it plays an important role in the appropriate timing of buying or selling stocks, which is one of the most popular and valuable areas in finance. In this work, we propose novel hybrid models for forecasting the one-time-step and multi-time-step close prices of DAX, DOW, and S&P500 indices by utilizing recurrent neural network (RNN)–based models; convolutional neural network-long short-term memory (CNN-LSTM), gated recurrent unit (GRU)-CNN, and ensemble models. We propose the averaging of the high and low prices of stock market indices as a novel feature. The experimental results confirmed that our models outperformed the traditional machine-learning models in 48.1% and 40.7% of the cases in terms of the mean squared error (MSE) and mean absolute error (MAE), respectively, in the case of one-time-step forecasting and 81.5% of the cases in terms of the MSE and MAE in the case of multi-time-step forecasting.