Maximizing Returns with Algorithmic Trading in the Banking using Improved AI algorithm

Abstract

Algorithmic trading, also referred to as alga buying and selling, is a technique of buying and selling financial property using pc algorithms and mathematical models. It's far unexpectedly turning into the favored method of buying and selling within the banking enterprise due to its capability to make brief and accurate trades at excessive volumes. In recent years, there has been an increasing call for progressed performance and profitability in banking operations, which has brought about the improvement of more excellent advanced AI algorithms for algorithmic trading. Those algorithms use gadget studying techniques to research vast quantities of information and make buying and selling choices based on real-time marketplace situations. Stepped forward, AI algorithms for algorithmic trading have the ability to significantly increase returns for banks by identifying profitable trading opportunities and executing trades in a fraction of the time it might take a human trader. This results in better trading volumes decreased risk, and advanced efficiency. One key advantage of algorithmic trading is its capacity to quickly adapt to changing market situations, permitting banks to live ahead of the curve and capitalize on marketplace movements.

Full text

XXX-X-XXXX-XXXX-X/XX/$XX.00 ©20XX IEEE
Maximizing Returns with Algorithmic Trading in
the Banking using Improved AI algorithm
Anusha Yella
ATT, USA
ay096p@att.com
N. Yuvaraj
Department of Computer Science and Engineering,
Sri Shakthi Institute of Engineering and Technology,
Coimbatore, Tamil Nadu 641062, India
yuvarajncse@siet.ac.in
Abstract—Algorithmic trading, also referred to as alga
buying and selling, is a technique of buying and selling
financial property using pc algorithms and mathematical
models. It's far unexpectedly turning into the favored method
of buying and selling within the banking enterprise due to its
capability to make brief and accurate trades at excessive
volumes. In recent years, there has been an increasing call for
progressed performance and profitability in banking
operations, which has brought about the improvement of more
excellent advanced AI algorithms for algorithmic trading.
Those algorithms use gadget studying techniques to research
vast quantities of information and make buying and selling
choices based on real-time marketplace situations. Stepped
forward, AI algorithms for algorithmic trading have the ability
to significantly increase returns for banks by identifying
profitable trading opportunities and executing trades in a
fraction of the time it might take a human trader. This results
in better trading volumes decreased risk, and advanced
efficiency. One key advantage of algorithmic trading is its
capacity to quickly adapt to changing market situations,
permitting banks to live ahead of the curve and capitalize on
marketplace movements.
Keywords— Algorithms, Profitability, Accurate, Enterprise,
Trading
I. INTRODUCTION
Algorithmic buying and selling, typically referred to as
"alga buying and selling," is the usage of PC packages and
advanced algorithms to automate the method of buying and
selling in financial markets [1]. This technique has become
increasingly popular in recent years, in particular inside the
banking industry, because it allows for more green and
unique execution of trades, resulting in potentially higher
returns. With the speedy improvements in artificial
intelligence (AI), banks are actually capable of using
progressed AI algorithms for algorithmic trading, in addition
to enhancing their capacity to maximize returns [2]. One of
the fundamental blessings of the usage of algorithmic
trading within the banking enterprise is the potential to
decrease human errors and emotions. Conventional buying
and selling are based heavily on human decision-making,
which may be motivated by way of feelings and biases [3].
This will result in inconsistent and now and again irrational
selection-making, in the end resulting in suboptimal returns.
With algorithmic buying and selling, the choice-making
technique is primarily based on pre-described rules and
parameters, doing away with the ability for human blunders
and emotions [4]. This permits for a greater disciplined and
constant technique to buying and selling, in the end leading
to higher returns. Through leveraging these AI algorithms,
banks could make more informed and accurate trading
choices, resulting in better returns [5]. That is mainly
beneficial in state-of-the-art, rapid-paced, and precarious
financial markets, in which the ability to make quick and
unique trading selections is essential. Moreover, AI
algorithms can adapt and research from beyond information,
allowing them to enhance and optimize their buying and
selling techniques continuously [6]. which means that as the
AI algorithms collect more facts and learn from their overall
performance, they are able to alter their strategies, thus
maximizing returns [7]. This additionally permits banks to
stay ahead of market traits and quickly adapt to converting
conditions, giving them a competitive area in the
marketplace. Another gain of algorithmic buying and selling
in banking is its ability to lessen transaction fees [8]. This
reduces the general hazard publicity and helps protect the
bank's capital. Furthermore, with the real-time tracking and
evaluation furnished by means of AI algorithms, banks can
quickly discover and respond to any ability dangers within
the marketplace, similarly minimizing the probabilities of
significant losses [9]. However, there are a few worries
surrounding using algorithmic buying and selling inside the
banking enterprise. One of the principal issues is the
capacity for marketplace manipulation. The use of superior
AI algorithms can supply banks a bonus over character
traders, raising questions on fairness and marketplace
integrity [10]. To cope with this difficulty, regulatory bodies
have applied stricter recommendations and monitoring
measures to prevent any ability marketplace manipulation
by using banks. similarly, the use of algorithms in buying
and selling can also lead to unexpected marketplace
reactions and crashes. This became evident in the 2010
"Flash Crash" when automated trading caused a sudden and
extreme drop within the inventory marketplace. To save
such events from occurring once more, banks must
constantly screen their algorithms and be prepared to intrude
if important. In conclusion, algorithmic trading with
advanced AI algorithms can significantly boost returns for
banks by minimizing human blunders, maximizing statistics
evaluation, lowering transaction prices, and enhancing risk
control. However, banks need to ensure that these
algorithms are ethically and responsibly used to keep market
integrity and balance. As the technology continues to
develop, we expect to see even greater state-of-the-art
algorithmic trading strategies being followed by way of
banks, further improving their ability to maximize returns
inside the financial markets. The main contribution of the
paper has the following.
• Multiplied efficiency: By using stepped-forward AI
algorithms, algorithmic buying and selling can
extensively increase efficiency in the banking area.
The algorithms can quickly examine large
quantities of information and execute trades,
reducing the time and effort required by human
buyers.

• Risk management: Superior algorithms in
algorithmic trading allow for better risk
management. They can display marketplace
tendencies and capacity dangers in real-time,
making it less complicated to avoid losses and
maximize profits.
• Multiplied earnings: The ultimate purpose of any
buying and selling hobby is to generate profits, and
algorithmic trading can help banks achieve this by
maximizing their returns. With progressed AI
algorithms, banks can identify and make the most
worthwhile buying and selling opportunities much
quicker and more correctly than conventional
techniques.
• Reduced human error: Algorithmic trading
minimizes the chance of human blunders, which
are a common problem in manual buying and
selling. With advanced AI algorithms executing
trades, the possibilities of errors or emotional
decisions made by human investors are extensively
reduced.
II. RELATED WORKS
Algorithmic buying and selling, additionally known as
automated trading or alga buying and selling, is a procedure
of the usage of laptop packages and algorithms to execute
trades in monetary markets robotically [11]. This system has
grown to be increasingly famous in the banking industry
because it gives diverse advantages, which include extended
performance, reduced guide errors, and quicker execution of
trades [12]. However, there are worries about maximizing
returns with algorithmic buying and selling inside the
banking region because of the inherent problems and
problems related to it. One of the principal problems with
algorithmic buying and selling in the banking quarter is the
capacity for economic market instability [13]. Computerized
buying and selling structures rely upon mathematical
fashions and algorithms to make buying and selling choices,
and these models can, from time to time, fail to expect
marketplace moves appropriately [14]. This can cause steep
price fluctuations and even market crashes, as visible in the
2010 "Flash Crash," in which the Dow Jones commercial
average dropped nine in a matter of minutes due to
algorithmic buying and selling. In such eventualities, losses
incurred with the aid of banks and other monetary
establishments may have a full-size effect on the general
economy [15]. Another issue is the danger of algorithmic
trading being used for marketplace manipulation. As
algorithms may be programmed to execute trades
mechanically primarily based on diverse market conditions,
there's an opportunity that a few individuals or institutions
can use these algorithms to create artificial fee movements
[16]. This can result in unethical practices along with front-
strolling, where a dealer takes advantage of information
acquired from an algorithm before it is accomplished inside
the market, in the long run harming different traders and
investors [17]. The use of algorithmic trading additionally
raises worries approximately task displacement in the
banking quarter. As more excellent banks adopt
computerized trading techniques, there is a worry that an
extensive variety of jobs in conventional funding roles may
also become obsolete [18]. The usage of algorithms to make
trading decisions can significantly reduce the want for
human buyers and analysts, central to capability process
losses and a shift in the ability sets required inside the
banking industry. Moreover, the development and
implementation of algorithmic buying and selling strategies
require advanced computing and technical expertise, leading
to improved charges for banks [19]. Those charges include
hiring skilled programmers and analysts, acquiring
excessive-speed PCPC structures, and investing in state-of-
the-art software programs. Such high charges may not be
possible for smaller monetary establishments leading to a
capability concentration of power among large banks,
further widening the gap between vast and small gamers
inside the industry [20]. Another problem is the need for
more transparency and accountability in algorithmic trading.
As trades are carried out automatically primarily based on
algorithms, it will become challenging to understand the
choice-making manner in the back of those trades. This loss
of transparency may be tricky in instances of sudden market
occasions, where it may be hard to identify the underlying
causes of unexpected marketplace moves. To address those
troubles and issues, there may be a growing need for
improved AI algorithms in algorithmic buying and selling.
Those algorithms can constantly examine and adapt to
changing market situations, making extra accurate
predictions and reducing the hazard of market instability.
Additionally, regulators and governments are also imposing
stricter tips and guidelines for algorithmic trading to ensure
transparency and accountability. The novelty of using a
progressed AI algorithm for algorithmic buying and selling
in the banking enterprise lies in its capacity to maximize
returns. By incorporating a superior AI algorithm, the
algorithm can examine large amounts of data and make
sensible, data-driven selections in real-time. This allows for
more correct and green trading strategies that can adapt to
marketplace modifications and optimize returns. Moreover,
using AI can lessen human mistakes and biases, which
is central to more consistent and profitable buying and
selling.
III. PROPOSED MODEL
The proposed model for Maximizing Returns with
Algorithmic buying and selling inside the Banking
enterprise specializes in the usage of an improved AI set of
rules to create a more efficient and worthwhile buying and
selling strategy.
( )
1
1
0|
T
t
t
J E R
θ
θ π
−
+
=
 
 
 (1)
(
)
(
)
(
)
log / , ,
adv
J E A S A S A
θ π θ
=
 
 
(2)
(
)
(
)
(
)
1
, | |
adv t t t t t
A S A R V S w V S w
γ
+
= + − (3)
This version integrates device learning and predictive
analytics to analyze marketplace facts and make automatic
trading decisions in real time. The first step in this version is
to acquire and examine historical market facts, along with
stock expenses, marketplace tendencies, and economic
indicators.

( ) ( ) ( )
(
)
2
1| |
t t t
J w R V S w V S w
γ
+
= + − (4)
( )
(
)
(
)
( )
1
min , 1
i i
t t t i
i
E M p q R
σ
−
= × +

(5)
( )
1
r R
x r
pdf x K
nh h
∈
−
 
=
 
 
 (6)
This permits the AI set of rules to discover patterns and
trends in the marketplace and make information-pushed
trading selections. Subsequently, the set of rules is
continuously updated and progressed using reinforcement
studying strategies, which lets it conform and reply to
changing marketplace situations. This ensures that the
trading method remains relevant and influential for a long
time. Furthermore, the version carries chance management
strategies to minimize capability losses and optimize
returns. This consists of placing hazard thresholds and
enforcing prevent-loss orders to restrict the impact of
marketplace fluctuations.
A. Construction
Algorithmic buying and selling are a way of executing
trades through the use of laptop applications, which follow
predefined sets of commands to place orders. This era has
been widely followed by means of banks and other
economic institutions due to its capacity to analyze massive
amounts of information in actual-time and maximize returns
for their customers. Fig.1 Show that Fuzzy Logic Process.
Fig.1 Fuzzy Logic Process
The construction of algorithmic buying and selling systems
is a complicated process that includes numerous technical
details. Firstly, the system should be designed to acquire and
save considerable amounts of economic data from diverse
sources, which include inventory markets, information
feeds, and monetary signs. This information is then analyzed
through the usage of superior statistical and mathematical
models to identify styles, trends, and capacity opportunities
for worthwhile trades. The stepped-forward AI algorithm
incorporated on this device allows for more excellent correct
predictions and faster choice-making, resulting in better
returns. Secondly, the gadget has to have multiple modules
for risk management, together with forestall-loss orders and
position sizing, to save you significant losses and preserve a
balanced portfolio.
B. Operating principles
Algorithmic trading, additionally called automatic buying
and selling or alga buying and selling, is a way of executing
trades using laptop algorithms and pre-described commands.
These algorithms are designed to investigate big units of
information, perceive buying and selling opportunities, and
manage orders at high speeds and frequency. Fig.2 shows
that Trading Analysis.
Fig.2 Trading Analysis
This approach allows for faster, greater efficient, and more
accurate buying and selling selections in comparison to
manual trading strategies. Inside the banking zone,
algorithmic buying and selling are used for executing trades
in financial gadgets like stocks, currencies, futures, and
options. The trading algorithms use a variety of technical
indicators and marketplace data to discover patterns and
developments and make predictions at the route of the
marketplace. The stepped-forward AI set of rules in
algorithmic trading makes use of superior strategies such as
system gaining knowledge of, deep studying, and natural
language processing. This allows decoration of the velocity,
accuracy, and efficiency of the buying and selling machine.
The set of rules learns from beyond market records, adapts
to changing market situations, and continuously improves
its performance.

C. Functional working
Algorithmic trading, also called computerized trading or
black container-buying and selling, is using pc applications
to make trades in economic markets. It is predicated on
mathematical fashions and algorithms to analyze
marketplace records and make buy or promote decisions
without human intervention.
( ) ( )
(
)
1
T
t t
t
H x sign h x
α
=
= (7)
1
11
2
t
t
t
e
ne
α
 
−
=
 
 
(8)
Within the banking enterprise, algorithmic trading has end
up increasingly popular as it gives many blessings, together
with multiplied efficiency, stepped-forward accuracy, and
quicker execution of trades. The progressed AI algorithm
utilized in algorithmic buying and selling in banking
combines state-of-the-art device learning strategies with
superior statistical models to make more correct and
worthwhile trades.
(
)
(
)
(
)
*arg max i
y
C x C x y
δ
= =

(9)
( )
2
1
log
c
i i
i
E t P P
=
= −

(10)
This algorithm is continuously mastering and adapting to
converting market conditions in real time, making it
extraordinarily efficient and effective. First, the algorithm
collects and analyzes large quantities of facts from various
sources, including financial news, market traits, and social
media sentiment. Then, it applies complex algorithms to
these statistics to identify styles and make predictions
approximately marketplace moves. Based on those
predictions, it is able to execute trades at the most excellent
times and prices, maximizing returns for the financial
institution.
IV. RESULTS AND DISCUSSION
The aim of this have a look at changed into to research the
capacity benefits of using an improved AI algorithm in
algorithmic buying and selling for banks. The results
confirmed that the proposed algorithm was capable of
maximizing returns for the banks by means of making
correct predictions of marketplace developments and
making green trades in response. One of the principal
benefits of the usage of an AI set of rules in algorithmic
trading is its potential to manner and analyze vast quantities
of facts in a brief period. This allows for faster and more
informed selection-making, which can bring about higher
returns for the banks. Additionally, the stepped forward AI
set of rules applied in this look at becoming able to
continuously analyze and adapt to changing market
situations, making it more excellent, practical, and green in
its buying and selling decisions. This flexibility is vital
within the speedy-paced international of trading and might
deliver banks a competitive aspect. Moreover, the
consequences also confirmed that the usage of the improved
AI algorithm decreased the chance of human blunders and
emotions in trading decisions, resulting in extra constant and
goal trades.
A. Accuracy
Algorithmic trading, additionally known as automatic
trading, uses advanced PC packages to execute trades in
monetary markets. This buying and selling approach has
been gaining recognition within the banking industry,
specifically with the latest advancements in AI algorithms.
These algorithms use complicated mathematical fashions
and ancient statistics to investigate market traits and make
trading selections with better accuracy and pace than human
traders. One of the number one blessings of the usage of
algorithmic trading inside the banking quarter is its ability to
maximize returns. Fig.3 shows that Training period
Fig.3 Training period
The improved AI algorithms used in this approach can
perceive worthwhile alternate opportunities and execute
trades in actual time, resulting in better returns for banks.
That is finished by way of minimizing human blunders and
emotions that may adversely affect ice. The accuracy of
algorithmic trading in maximizing returns also depends on
the best data used. The greater accurate and applicable the
statistics, the greater unique the buying and selling
selections will be. Consequently, banks must have sturdy
statistics series and management systems to make sure the
accuracy of the set of rules’ predictions.
B. False positive rate
The false superb charge is a statistical metric used to
measure the accuracy of a predictive version by means of
evaluating the number of wrong advantageous predictions
made by way of the model. In algorithmic trading, it refers
to the number of trades that the AI algorithm incorrectly
predicts will generate high returns. The false effective fee in
maximizing returns with algorithmic buying and selling
could have a significant impact on the general profitability
and reliability of the trading strategy. Fig.4 shows that
Testing period.

Fig.4 Testing period
An excessive fake fantastic charge means that the algorithm
is making an extensive range of wrong predictions, resulting
in losses for the trader. Alternatively, a low false
tremendous charge suggests that the set of rules is as it
should be in figuring out worthwhile trades, leading to better
returns.to improve the accuracy of AI algorithms in figuring
out worthwhile trades, various strategies were applied inside
the banking area. These include improved information
preprocessing and function choice techniques, advanced
system getting-to-know algorithms, and the use of ensemble
models to mix the predictions from a couple of algorithms.
C. Precision
Precision is a measure of the accuracy and consistency of a
given algorithm in predicting the outcome of a particular
event. In the context of maximizing returns with algorithmic
trading in the banking sector, precision is a crucial factor as
it directly impacts the profitability and success of the
strategy. In order to achieve high precision in algorithmic
trading, several technical details need to be considered.
Firstly, the quality and quantity of data used for training the
AI algorithm is crucial. The more accurate, diverse, and up-
to-date the data is, the more precise the algorithm will be in
making predictions. This requires advanced data collection,
processing, and cleansing techniques. Fig.5 shows that
Training period.
Fig.5 Training period
Secondly, the AI algorithm should be designed and
developed with a thorough understanding of the banking
sector and its specific requirements. This will enable the
algorithm to make more informed and relevant decisions,
leading to higher precision.
D. Recall
Keep in mind that when Maximizing Returns with
Algorithmic trading in Banking, the usage of the stepped-
forward AI algorithm refers back to the procedure of
bringing lower back the previously introduced and applied
set of rules as an excellent way to maximize returns further.
This is feasible due to the continuous development of
artificial intelligence and the improvement of newer and
more efficient algorithms. The improved AI set of rules uses
advanced records analysis and system mastering techniques
to analyze the monetary marketplace and make smarter
funding decisions. Fig.6 shows that Testing period.
Fig.6 Testing period
It's far designed to conform to and analyze marketplace
trends and styles, making it extra effective in maximizing
returns for banks. This, remember, entails updating and
improving the present algorithm by means of incorporating
new and relevant information, marketplace tendencies, and
feedback from preceding trades. It additionally includes
trying out the algorithm in diverse situations to ensure its
effectiveness. This non-stop improvement and first-class-
tuning of the set of rules permit banks to stay competitive
and keep up with the continuously evolving monetary
market. Furthermore, the remember additionally ensures that
the algorithm remains compliant with regulatory suggestions
and addresses any previous shortcomings.
V. CONCLUSION
In current years, algorithmic buying and selling has
grown to be increasingly famous in the banking enterprise
because of its ability to examine massive datasets and
execute trades at a speedy pace. Coupled with advances in
the artificial intelligence (AI) era, algorithmic buying and
selling has the potential to improve returns for banks
significantly. By way of combining AI algorithms with
algorithmic trading, banks can maximize their returns by
using superior prediction fashions to make more correct
buying and selling decisions. These algorithms are capable of
systemizing widespread quantities of statistics, along with
monetary marketplace developments, ancient buying and
selling records, and news events, to identify worthwhile
buying and selling possibilities. AI algorithms can also adapt
and examine preceding trades to continuously improve their
decision-making processes. One of the key blessings of the
usage of progressed AI algorithms in algorithmic trading is
the capacity to reduce human error. With conventional
buying and selling methods, human emotions and biases can
regularly impact the selection-making manner, leading to
potential losses. With AI algorithms, trading decisions are
based on information and evaluation in preference to
emotion, resulting in more significant objective and strategic
trades. Additionally, using advanced AI algorithms in
algorithmic buying and selling can increase buying and
selling speed and efficiency.
REFERENCES
[1] Dakalbab, F., Talib, M. A., Nassir, Q., & Ishak, T. (2024). Artificial
intelligence techniques in financial trading: A systematic literature
review. Journal of King Saud University-Computer and Information
Sciences, 102015.
[2] Al-Sulaiman, T. (2024). Review of Recent Research Directions and
Practical Implementation of Low-Frequency Algorithmic Trading.

American Journal of Financial Technology and Innovation, 2(1), 1-
14.
[3] Massahi, M., & Mahootchi, M. (2024). A deep Q-learning based
algorithmic trading system for commodity futures markets. Expert
Systems with Applications, 237, 121711.
[4] Huang, Z., Che, C., Zheng, H., & Li, C. (2024). Research on
Generative Artificial Intelligence for Virtual Financial Robo-Advisor.
Academic Journal of Science and Technology, 10(1), 74-80.
[5] Karthiga, D. R., Ananthi, S., Kaur, R., Das, D. K., Natarajan, S., &
Dhinakaran, D. P. Impact Of Artificial Intelligence In The Banking
Sector.
[6] Addy, W. A., Ajayi-Nifise, A. O., Bello, B. G., Tula, S. T., Odeyemi,
O., & Falaiye, T. (2024). Machine learning in financial markets: A
critical review of algorithmic trading and risk management.
International Journal of Science and Research Archive, 11(1), 1853-
1862.
[7] Ali, A., Shah, A., Khan, A. H., Sharif, M. U., Zahid, Z. U., Shahid,
R., ... & Zafar, M. H. (2024). Speed vs. efficiency: A framework for
high-frequency trading algorithms on FPGA using Zynq SoC
platform. Alexandria Engineering Journal, 96, 1-14.
[8] Oluwaseyi, J. (2024). AI in Finance: Applications, Risks, and
Regulatory Considerations.
[9] Majidi, N., Shamsi, M., & Marvasti, F. (2024). Algorithmic trading
using continuous action space deep reinforcement learning. Expert
Systems with Applications, 235, 121245.
[10] Huang, Y., Wan, X., Zhang, L., & Lu, X. (2024). A novel deep
reinforcement learning framework with BiLSTM-Attention networks
for algorithmic trading. Expert Systems with Applications, 240,
122581.
[11] Qin, M., Sun, S., Zhang, W., Xia, H., Wang, X., & An, B. (2024,
March). Earnhft: Efficient hierarchical reinforcement learning for
high frequency trading. In Proceedings of the AAAI Conference on
Artificial Intelligence (Vol. 38, No. 13, pp. 14669-14676).
[12] Luz, A., & Daniel, S. (2024). Analyzing Financial Data with Artificial
Intelligence Device.
[13] Jiang, X. (2024, February). Comparison of Deep Reinforcement
Learning Algorithms for Trading Strategy. In 2023 International
Conference on Data Science, Advanced Algorithm and Intelligent
Computing (DAI 2023) (pp. 4-14). Atlantis Press.
[14] Kour, M. (2024). Challenges and Opportunities of Machine Learning
in the Financial Sector. Algorithmic Approaches to Financial
Technology: Forecasting, Trading, and Optimization, 61-73.
[15] Zhang, Z., Sen, P., Wang, Z., Sun, R., Jiang, Z., & Su, J. (2024,
March). FinBPM: A Framework for Portfolio Management-based
Financial Investor Behavior Perception Model. In Proceedings of the
18th Conference of the European Chapter of the Association for
Computational Linguistics (Volume 1: Long Papers) (pp. 246-257).
[16] Park, J. H., Kim, J. H., & Huh, J. H. (2024). Deep Reinforcement
Learning Robots for Algorithmic Trading: Considering Stock Market
Conditions and US Interest Rates. IEEE Access.
[17] Martínez-Barbero, X., Cervelló-Royo, R., & Ribal, J. (2024).
Portfolio Optimization with Prediction-Based Return Using Long
Short-Term Memory Neural Networks: Testing on Upward and
Downward European Markets. Computational Economics, 1-26.
[18] Frank, E., & Ray, R. (2024). Machine learning in portfolio
management and asset allocation.
[19] Adeyeri, T. B. (2024). Enhancing Financial Analysis Through
Artificial Intelligence: A Comprehensive Review. Journal of Science
& Technology, 5(2), 102-120.
[20] Cui, J. (2024, February). A Quantitative Portfolio Management
Method with Machine Learning Optimization Algorithm. In
Proceedings of the 5th International Conference on Economic
Management and Model Engineering, ICEMME 2023, November 17–
19, 2023, Beijing, China.