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.

Key-Words: - Decision support systems, stock investment, portfolio optimization, modern portfolio theory.

Received: August 29, 2022. Revised: October 2, 2023. Accepted: October 13, 2023. Published: November 3, 2023.

which stocks will be chosen and how many shares

will be allocated to these stocks. The financial value

of the portfolio depends on the prices of the stocks it

contains and can be continuously monitored by the

investor. Portfolio optimization usually specifies the

Researchers usually try to solve this confusion using

computer applications or mathematical models.

However, the variety of investment strategies and

the intricacy of investment choices have consistently

grown over the past two decades. This rise has

DSS helps decision-makers in the decision-making

process with its features, such as analyzing data by

processing, modeling decisions in different

scenarios, and predicting the consequences of future

actions. These systems can collect data from various

sources, such as business data, financial

information, and market research results, and

analyze this data to provide accurate information to

decision-makers. DSS helps managers or investors

make faster, more effective, and more informed

and strategy optimization, [5]. Therefore, DSS is a

helpful mechanism for choosing the most suitable

portfolio according to the investor's sector among

different portfolios. These systems determine an

optimal asset allocation, considering an investor's

risk tolerance, return targets, and other constraints.

This type of optimization often requires statistical

analysis of the returns and risks of assets in a

particular portfolio. Using these systems can

increase an investor's ability to maximize potential

optimal portfolio for stock investments with an

overview and present DSS studies aimed at

optimizing financial decisions in this context.

Section 3 outlines the objectives, methodologies,

and advantages of DSS structures employed in

portfolio optimization within the context of stock

investments. Moreover, the pivotal role of DSS in

formulating and managing stock investment

portfolios is elaborated. DSS structures elucidate

how researchers and practitioners can make more

[28], the authors significantly developed the theory

of portfolio selection and management by

considering various factors that investors face in the

real world. They have integrated real-life constraints

into the model, such as the proportion of a portfolio

and, [32], the authors developed the short-sale

theme further. The single-index model and the

Multi-Index Models address the complications

brought by the increasing number of securities in

determining the expected return and variance of

optimal portfolios, [24], [33]. The Single Index

Model developed by William Sharpe is a

methodology employed to gauge the performance of

risky assets within a portfolio. This model facilitates

portfolio managers in quantifying the returns

manager, [34]. A mathematical model for the

portfolio optimization problem was presented by the

authors of the Black-Scholes model, which is

famous for option pricing, [35].

determining how to allocate their assets to maximize

expected returns while minimizing risk. Monte

Carlo simulations are commonly employed in

solving these optimization problems. The Monte

Carlo simulation is a statistical computation method

assigned randomly, and the process is repeated. This

gives investors valuable insights into how their

portfolios will perform under different scenarios,

equipping them to make more informed investment

decisions, [36]. Monte Carlo simulation methods

continue to be developed in various DSS using other

programming languages today, [22], [37], [38], [39],

[40].

rise to the Financial Asset Pricing Model, a

tolerance, return expectations, and other criteria.

Palma-dos-Reis and Zahedi (1999) have used DSS

to explore and investigate the influence of investors'

personal traits on their use of models when making

investment decisions, [5]. In, [54], it is utilized the

for portfolio optimization problems. The authors

propose a multi-criteria decision-making model

based on the PROMETHEE method to select

superior stocks in the stock market and apply this

model to the Tehran Stock Exchange as a real-world

case in, [57]. As a stock selection strategy, it is

developed two DEA models for the Taiwan Stock

Exchange in, [58], while it is implemented a

prototype named MISMIS for the Hang Seng Index

in, [59]. The DSS introduced in, [60], is a design

intelligently recognizes trading signals and aids

investors in their decisions, utilizing techniques like

neural networks, particle swarm optimization, and

denoising. Simulations showed that traders could

achieve higher returns using this decision support

system, highlighting the advantages of incorporating

adaptive and intelligent decision-making into

forecasts. In the literature, many studies have been

conducted using multi-criteria decision-making

techniques such as AHP (Analytic Hierarchy

as Convolutional Neural Networks (CNNs),

Artificial Neural Networks (ANNs), Recurrent

Neural Networks (RNNs), machine learning, and

genetic algorithms have been applied to complex

problems in financial markets, such as stock market

investment. Thanks to their capabilities in

processing large amounts of data and learning, these

techniques have been used in predicting stock

prices, portfolio management, and risk analysis.

mathematical models to maximize the potential of

portfolio management. In the following, we describe

some prominent DSS found in the literature and

review which methods and how these systems are

used.

influence the use of the DSS and model selection.

However, the generalizability of these results could

be improved, and further experiments with a larger

sample size are planned. The research presents

significant initial steps on how a financial DSS can

be personalized based on user characteristics.

combines decision analysis with traditional

investment evaluation procedures to assist decision-

makers in making better investment decisions, [54].

considering the decision maker's risk stance and

preferences. Although the system has been tested on

the Shanghai Stock Exchange, there is a need for

extensive testing and evaluation of its adaptability to

other stock markets.

knowledge exchange framework focused on how

agents can exchange information and knowledge to

solve common problems collaboratively.

Specifically designed for stock trading, the Multi-

algorithms to make the decision-making process in

portfolio management easier and faster. This model

presents investors with different scenarios derived

from complex data. The decision support system,

MISMIS was developed for financial data

predictions, [59]. This system allows investors to

create complex prediction models in a step-by-step

manner. Results from each model can be used as

inputs for the next one. This multi-level approach

optimize their portfolios, evaluate potential risks,

and predict the possible returns of different

investment scenarios.

large-scale predictive decision-making approach for

the CLUS-MCDA algorithm was expanded for the

asset allocation phase. An experiment analyzing

over 70 cryptocurrencies was conducted to verify

the method's reliability, The results indicate that the

proposed decision support system is a trustworthy

tool for real-world big data challenges.

Recommendations include verifying time series

forecasting models, linking to a real-time database,

incorporating natural language processing and

semantic analysis into the methodology, and

structures. These techniques enable investors to

apply a scientific and systematic approach to

complex investment decisions by considering

different criteria and constraints. Specifically,

decision-makers prefer the Analytic Hierarchy

Process in the prioritization process between

criteria; they use TOPSIS to determine the

proximity of alternative stocks to the ideal solution,

and they favor Vikor to reveal optimal compromise

solutions. Methods like ELECTRE and

utilized for the DSS structures used in stock

portfolio optimization. Investors frequently prefer

Microsoft Excel and its additional Solver tool for

basic optimization tasks. MATLAB is employed to

implement complex algorithms, while Python,

enriched with libraries such as pandas and numpy, is

becoming increasingly popular for financial analysis

and optimization. The R language stands out for

financial analysis with "Performance Analytics"

portfolio optimization processes. These criteria

allow investors to consider their social and

environmental impacts while maximizing financial

returns.

[21] Tükel, G. Ö. & Önem, H. B. (2020). Solving a

[23] Tobin, J. (1958). Estimation of relationships

[24] Sharpe, W. F. (1963). A simplified model for

[25] Sharpe, W. F. (1964). Capital asset prices: A

[27] Budgets, C., & Lintner, J. (1965). The

[28] Mossin, J. (1966). Equilibrium in a capital

[29] Brennan, M. J. (1971). Capital market

[31] Levy, H. (1983). The capital asset pricing

[32] Schnabel, J. A. (1984). On mean-variance

[33] Perold, A. F. (1984). Large-scale portfolio

[34] Fama, E. F., & French, K. R. (1993).

[44] Black, F., & Litterman, R. (1992). Global

[47] Müjdecı, S. Y., Alnahas, D., Öcal, M., Duru,

[48] Stoilov, T., Stoilova, K., & Vladimirov, M.

[50] Herlansyah, R. R., & Saepudin, D. (2023).

[53] Eom, S., & Kim, E. (2006). A survey of

[54] Poh, K. L. (2000). An intelligent decision