CHAPTER 1

The Role of Computational Modeling in Modern Finance

K. Pallavi*, Dr. Sonal Trivedi
VIT Bhopal University, Kothri Kalan, Near Sehore (M.P.)
Abstract: This paper explores the application of computational modeling in finance,
highlighting its significance in predictive analytics, risk management, and algorithmic
trading. By reviewing various models and techniques, the paper illustrates how
computational finance can enhance decision-making processes and optimize financial
strategies. The study also discusses challenges and future directions, providing a
comprehensive understanding of the field's current state and potential developments.
Keywords: Computational Modelling, Modern Finance, Predictive analytics, risk
management, algorithmic trading.

INTRODUCTION
Computational modeling has emerged as a cornerstone of modern finance, reflecting
the sector's growing reliance on advanced mathematical and algorithmic techniques to
navigate the complexities of global financial markets. This evolution is driven by the
increasing need for precision, efficiency, and speed in financial decision-making
processes, areas where traditional methods often fall short. The origins of
computational modeling in finance can be traced back to the development of
foundational models such as the Black-Scholes model for option pricing in the early
1970s. Since then, the field has undergone significant transformations, incorporating
innovations in computer science, statistics, and artificial intelligence (Groesser, S. N.,
& Jovy, N. 2016). These advancements have expanded the scope and capabilities of
computational models, enabling them to address a broader range of financial problems
with greater accuracy and sophistication.
In contemporary finance, computational models are utilized for a variety of purposes,
including predictive analytics, risk management, and algorithmic trading. Predictive
models leverage machine learning and AI to forecast market trends and asset prices,
providing investors and institutions with valuable insights. Risk management models
quantify and mitigate potential financial risks, ensuring stability and compliance with
regulatory standards. Algorithmic trading, driven by complex algorithms and high-
frequency data analysis, has revolutionized trading strategies, enhancing market
efficiency and liquidity. Despite their numerous advantages, computational models in
finance are not without challenges. Issues such as data quality, computational
limitations, and model risk necessitate continuous validation and refinement of these
models. Additionally, the increasing use of AI and machine learning raises ethical and
regulatory concerns that must be addressed to ensure fairness, transparency, and
accountability in financial practices.
This paper aims to provide a comprehensive overview of the role of computational
modeling in modern finance. It will explore the various types of computational
models, their applications in predictive analytics, risk management, and algorithmic
trading, and the challenges and limitations associated with their use (Cavalcante et.al.,
2016). Furthermore, the paper will discuss future directions in computational finance,
highlighting emerging trends and potential advancements that could shape the
industry's future. By examining these aspects, the paper seeks to illustrate how
computational modeling can enhance financial decision-making processes, optimize
strategies, and ultimately contribute to a more robust and resilient financial system.
*Corresponding author K.Pallavi: VIT Bhopal University, Kothri Kalan, Near Sehore (M.P.); E-mail:
pallavi.k2022@vitbhopal.ac.in
 The chapter aims to fulfil the following objectives:
- To review the types of computational models used in finance.
- To examine the role of these models in predictive analytics, risk management,
and algorithmic trading.
- To discuss the challenges and limitations associated with computational
finance.
- To explore future trends and potential advancements in the field.

2. Overview of Computational Models in Finance

2.1 Definition and Types of Computational Models:
Computational models in finance refer to mathematical and algorithmic frameworks designed to
simulate financial markets, forecast economic variables, and optimize investment strategies. These
models can be broadly categorized into stochastic models, agent-based models, and machine-
learning models.

2.2 Historical Evolution and Advancements:

The use of computational models in finance dates back to the mid-20th century with the
development of the Black-Scholes model for option pricing (Amman et.al., 1996). Since then, the
field has witnessed significant advancements, including the integration of machine learning and
artificial intelligence (AI) techniques, which have enhanced the accuracy and efficiency of financial
models.

3. Predictive Analytics in Finance

Predictive analytics in finance leverages historical data and advanced algorithms to forecast future
financial trends, prices, and economic behaviors. This section delves deeper into the techniques and
applications of machine learning (ML) and artificial intelligence (AI) in financial forecasting,
providing a comprehensive understanding of their significance, methodologies, and evaluation
metrics (Tesfatsion, 2023).

3.1 Machine Learning and AI in Financial Forecasting

Machine learning and AI have become pivotal in financial forecasting, thanks to their ability to
process vast amounts of data and identify complex patterns that traditional statistical methods may
overlook. Here are some key techniques and their applications:

Neural Networks:
- Overview: Neural networks, inspired by the human brain's structure, consist of
interconnected nodes (neurons) that process data in layers. They are particularly effective in
capturing non-linear relationships.
- Application: In stock price prediction, neural networks analyze historical price data,
volume, and other relevant features to forecast future prices. Advanced architectures like Long
Short-Term Memory (LSTM) networks are used for time-series forecasting, capturing temporal
dependencies in sequential data.

Support Vector Machines (SVM):

- Overview: SVMs are supervised learning models used for classification and regression
tasks. They work by finding the optimal hyperplane that separates data points of
different classes with the maximum margin.
- Application: SVMs are employed in market trend analysis, classifying market
conditions (e.g., bull or bear markets) based on historical data and technical indicators. This helps
traders and investors make informed decisions about entry and exit points.

Random Forests:
- Overview: Random forests are ensemble learning methods that build multiple
decision trees and aggregate their predictions. This reduces overfitting and improves model
robustness.
- Application: In credit scoring, random forests analyze borrower data (e.g., income, credit
history, loan amount) to predict the likelihood of default, aiding in credit risk assessment.

Deep Learning:
- Overview: Deep learning, a subset of machine learning, involves neural networks with
multiple hidden layers. These models can automatically extract high-level features from raw data.
- Application: Deep learning models, such as convolutional neural networks (CNNs), are
used for sentiment analysis by processing textual data from news articles, social media, and
financial reports to gauge market sentiment and its impact on stock prices.

3.2 Case Studies

Stock Price Prediction:

- Example: A study used LSTM networks to predict stock prices of major companies. The
model was trained on historical price data, technical indicators, and market news
sentiment. Results showed that the LSTM network outperformed traditional ARIMA
models in terms of prediction accuracy and robustness (Chen & Liao, 2005).

Market Trend Analysis:

- Example: Researchers applied SVMs to classify market conditions based on historical price
data and technical indicators like moving averages and relative strength index (RSI) (Agarwal,
2019). The SVM model provided reliable signals for market entry and exit, helping traders
optimize their strategies.

Credit Risk Assessment:

- Example: A financial institution implemented a random forest model to predict the
probability of loan default. By analyzing borrower characteristics, the model improved the accuracy
of credit risk assessments, enabling better decision-making in loan approvals.
 Mesoscopic Simulation of Rarefied Gas STP Flows on Chemical and Biomedical Engineering 5

3.3 Evaluation of Predictive Models

To ensure the reliability and effectiveness of predictive models, various evaluation metrics
and methods are employed:

Mean Absolute Error (MAE):

- Definition: MAE measures the average magnitude of errors between predicted and
actual values, without considering their direction.
- Formula: \( \text{MAE} = \frac{1}{n} \sum_{i=1}^{n} | y_i - \hat{y}_i | \)
- Usage: Lower MAE values indicate better predictive accuracy. It is widely used in
stock price prediction and regression tasks.

Root Mean Square Error (RMSE):

- Definition: RMSE measures the square root of the average squared differences
between predicted and actual values, penalizing larger errors more heavily.
- Formula: \( \text{RMSE} = \sqrt{\frac{1}{n} \sum_{i=1}^{n} (y_i - \hat{y}_i)^2} \)
- Usage: RMSE is useful for assessing model performance in forecasting tasks,
where larger errors need to be penalized.

R-squared (Coefficient of Determination):

- Definition: R-squared indicates the proportion of variance in the dependent variable
explained by the model.
- Formula: \( R^2 = 1 - \frac{\sum_{i=1}^{n} (y_i - \haty_i)^2}{\sum_{i=1}^{n} (y_i -
\bar{y})^2} \)
- Usage: Higher R-squared values signify better model fit. It is commonly used in
regression analysis to evaluate the explanatory power of predictive models.

Confusion Matrix:
- Definition: A confusion matrix is a table used to evaluate the performance of
classification models, showing the true positives, true negatives, false positives, and false
negatives.
- Usage: It is crucial in assessing the accuracy, precision, recall, and F1-score of
classification models used in credit risk assessment and market trend analysis.

Cross-Validation:
- Definition: Cross-validation involves partitioning the dataset into multiple subsets,
training the model on some subsets and validating it on others. This process is repeated
several times to ensure robust performance.
- Usage: Cross-validation helps in preventing overfitting and assessing the
generalizability of the predictive model.

Backtesting:
- Definition: Backtesting involves applying the predictive model to historical data to
evaluate its performance in real-world scenarios.
- Usage: It is essential in financial forecasting and algorithmic trading to validate
model effectiveness and reliability before deployment.

Predictive analytics in finance, powered by machine learning and AI, has revolutionized
how financial institutions forecast market trends, assess risks, and make investment
decisions (Focardi & Fabozzi, 2004). By leveraging advanced algorithms and robust
evaluation metrics, these models provide valuable insights, enabling more
informed and strategic financial practices.

4. Risk Management

Risk management is a critical function within the financial industry, aiming to identify,
assess, and mitigate potential risks that can adversely affect an organization's financial
health. Computational models play an indispensable role in this process, providing
quantitative tools and techniques to evaluate and manage various types of financial risks,
including market risk, credit risk, operational risk, and liquidity risk.

4.1 Role of Computational Models in Risk Assessment:

Computational models are essential in quantifying and managing financial risks,

enabling institutions to make informed decisions and comply with regulatory standards.
These models use historical data and statistical methods to estimate potential losses and
identify risk factors. By simulating different scenarios and stress-testing portfolios,
financial institutions can better understand the implications of adverse market
conditions and develop strategies to mitigate these risks (Axtell & Farmer, 2022). This
proactive approach helps in maintaining financial stability and protecting stakeholders'
interests.

4.2 Value at Risk (VaR):

Value at Risk (VaR) is one of the most widely used metrics for risk assessment in the
financial industry. It estimates the maximum potential loss of a portfolio over a specified
period at a given confidence level. For instance, a one-day VaR at the 95%
confidence level means there is a 5% chance that the portfolio will lose more than the VaR
amount in one day. VaR can be calculated using various methods, including historical
simulation, variance-covariance, and Monte Carlo simulation. Historical simulation
involves using past market data to simulate future price movements, while the variance-
covariance method assumes normal distribution of returns and uses the mean and standard
deviation to estimate VaR (Yerashenia, & Bolotov, 2019). Monte Carlo simulation, on the
other hand, generates a large number of random price scenarios to estimate potential losses.
Each method has its strengths and weaknesses, and the choice of method depends on the
specific requirements and risk profile of the institution.

4.3 Monte Carlo Simulations:

Monte Carlo simulations are a powerful tool in risk management, particularly for complex
portfolios and instruments with non-linear payoffs. These simulations involve generating a
large number of random scenarios for market variables such as interest rates, stock prices,
and exchange rates. By running the model under these scenarios, financial institutions can
obtain a probabilistic distribution of potential outcomes, providing a comprehensive view
of the risks involved (Veryzhenko, 2012). Monte Carlo simulations are especially useful
for assessing the risk of derivatives and structured products, where traditional analytical
methods may fall short. They help in identifying the tail risks and understanding the impact
of extreme market movements, enabling institutions to develop robust risk mitigation
strategies.

4.4 Real-World Applications:

Computational models are extensively used in various real-world applications of risk
management, providing valuable insights and aiding in decision-making processes. In
credit risk management, for instance, models such as credit scoring and default prediction
are used to assess the creditworthiness of borrowers. These models analyze various factors,
including the borrower’s credit history, income, and loan characteristics, to estimate the
probability of default. Financial institutions use these predictions to set credit limits,
determine interest rates, and make lending decisions.

In portfolio management, optimization models like the Markowitz mean-variance model

help in constructing portfolios that maximize returns for a given level of risk. By
considering the expected returns, variances, and covariances of different assets, these
models identify the optimal asset allocation that achieves the desired risk-return trade- off.
Stress testing and scenario analysis are also integral parts of portfolio risk management,
allowing institutions to evaluate the impact of adverse market conditions on portfolio
performance.

In addition to market and credit risk, computational models are used to manage operational
risk, which arises from inadequate or failed internal processes, systems, and controls. For
example, models can simulate the potential losses from cyber- attacks, fraud, or regulatory
breaches, helping institutions to implement effective risk controls and contingency plans.

Computational models in risk management offer a systematic and quantitative approach to

identifying, assessing, and mitigating financial risks. By leveraging advanced statistical
methods and simulations, these models provide valuable insights into potential losses and
help financial institutions to develop robust risk management strategies. As the financial
landscape continues to evolve, the importance of computational models in ensuring
financial stability and resilience will only grow.

5. Algorithmic Trading

Algorithmic trading, also known as algo trading or automated trading, involves the use of
computer algorithms to automatically execute trades based on predefined criteria, without
the need for human intervention. This technology has revolutionized financial markets by
increasing trading efficiency, reducing transaction costs, and enabling the execution of
complex trading strategies at speeds and frequencies beyond human capability. The
algorithms used in trading are designed to analyze market conditions, identify trading
opportunities, and execute orders at optimal prices, often within fractions of a second.

5.1 Introduction to Algorithmic Trading:

Algorithmic trading systems are built on advanced computational techniques that

leverage historical and real-time market data to make trading decisions. These systems can
process vast amounts of information, including price movements, trading volumes, and
news feeds, to generate buy or sell signals. The primary advantages of algorithmic trading
include the ability to execute high-frequency trades with precision, minimize human error,
and exploit market inefficiencies that are often too subtle for manual detection (Zenios,
1999). As a result, algorithmic trading has become a dominant force in global financial
markets, particularly in equity, foreign exchange, and commodities trading.

5.2 Computational Techniques in Developing Trading Algorithms:

Various computational techniques are employed to develop robust and effective trading
algorithms. These include:

- Statistical Arbitrage: This strategy involves exploiting price discrepancies between

related financial instruments. Algorithms identify mispricings by analyzing historical price
relationships and statistical properties. When a deviation from the expected relationship is
detected, the algorithm executes trades to profit from the eventual convergence of prices.
For example, pairs trading is a common form of statistical arbitrage where two correlated
stocks are traded based on their relative price
movements.

- Trend Following: Trend-following algorithms aim to capitalize on market

momentum by identifying and following established price trends. These algorithms use
technical indicators, such as moving averages and momentum oscillators, to generate
trading signals. When a security's price breaks above a moving average, the algorithm may
initiate a buy order, expecting the upward trend to continue. Conversely, a break below the
moving average may trigger a sell order. Trend-following strategies are popular in both
short-term and long-term trading.

- Market Making: Market-making algorithms provide liquidity to the market by

continuously placing buy and sell orders. These algorithms profit from the bid-ask spread,
which is the difference between the buying and selling prices. Market makers aim to
maintain a neutral position by adjusting their quotes based on market conditions and order
flow. This strategy requires sophisticated risk management techniques to minimize
exposure to adverse price movements and ensure profitability.

- Sentiment Analysis: Sentiment analysis algorithms assess market sentiment by

analyzing textual data from news articles, social media, and financial reports. Natural
language processing (NLP) techniques are used to gauge the overall sentiment and its
potential impact on asset prices. Positive sentiment may lead to buy signals, while negative
sentiment may trigger sell signals (Melnik, R. (Ed.). 2015). These algorithms help traders
anticipate market reactions to news events and adjust their positions accordingly.

5.3 Impact on Market Dynamics:

Algorithmic trading has had a profound impact on market dynamics, influencing liquidity,
volatility, and price discovery. One of the key benefits of algorithmic trading is the
enhancement of market liquidity. By providing continuous bid and ask quotes, market-
making algorithms narrow the bid-ask spreads, making it easier for traders to enter and exit
positions. This increased liquidity contributes to more efficient markets and better price
discovery.

However, algorithmic trading also introduces risks and challenges. High-frequency trading
(HFT), a subset of algorithmic trading, involves executing a large number of orders at
extremely high speeds. While HFT can improve market efficiency, it can also contribute to
market volatility. Flash crashes, where prices plummet rapidly due to a cascade of
algorithmic trades, are a notable risk associated with HFT. These events highlight the need
for robust risk management systems and regulatory oversight to ensure market stability.

Furthermore, the proliferation of algorithmic trading raises concerns about market

manipulation and fairness. Strategies such as spoofing (placing and then canceling large
orders to manipulate prices) and front-running (executing orders based on advanced
knowledge of pending trades) can distort market prices and disadvantage other market
participants (Hilber et.al., 2013).
Regulators and exchanges are continuously working to address these issues by
implementing stricter rules and surveillance mechanisms.

Algorithmic trading represents a significant advancement in the financial industry, offering

numerous benefits in terms of efficiency, cost reduction, and market access. By utilizing
sophisticated computational techniques, trading algorithms can analyze vast amounts of
data, execute trades with precision, and adapt to changing market conditions. However, the
impact of algorithmic trading on market dynamics also necessitates careful risk
management and regulatory oversight to mitigate potential risks and ensure fair and orderly
markets. As technology continues to evolve, algorithmic trading will likely play an even
more prominent role in shaping the future of financial markets.

6. Challenges and Limitations

Despite the transformative potential of computational models in finance, their

implementation and effectiveness are accompanied by several challenges and
limitations. Addressing these issues is crucial for optimizing model performance and
ensuring their reliability in financial decision-making.

6.1 Computational and Data Limitations:

Computational models heavily rely on the availability and quality of data. One significant
challenge is ensuring data accuracy and completeness. Financial data can be noisy,
incomplete, or subject to errors, which can adversely affect model performance and lead to
misleading conclusions. For instance, inaccurate historical data may result in flawed
predictions or risk assessments. Furthermore, computational models often require
substantial processing power, especially for complex simulations or high-frequency trading
algorithms. Limited computational resources can constrain the ability to run large-scale
models or perform real-time analysis, potentially impeding the timeliness and accuracy of
decision-making.

6.2 Model Risk and Validation:

Model risk arises when a computational model fails to accurately represent the financial
phenomena it is designed to predict. This risk can stem from incorrect assumptions,
oversimplified representations of complex market dynamics, or inappropriate model
specifications. For example, a model that assumes normal distribution of returns might not
capture extreme market events, leading to underestimation of risk. To mitigate model risk,
rigorous validation and backtesting are essential. This involves testing the model on
historical data and different market conditions to ensure its robustness and reliability.
However, backtesting alone cannot guarantee future performance, as market conditions
evolve and new risk factors emerge (Lai & Xing, 2008). Thus, ongoing model validation
and adjustment are necessary to maintain accuracy and relevance.

6.3 Ethical Considerations and Regulatory Challenges:

The rise of algorithmic trading and AI-driven financial models introduces ethical and
regulatory concerns. Algorithmic trading, for instance, can lead to market manipulation or
unfair advantages if not properly regulated. Strategies such as spoofing or quote stuffing
can distort market prices and harm other participants. Additionally, the use of AI in
financial decision-making raises questions about transparency and accountability. AI
models often operate as "black boxes," making it difficult to understand or explain their
decision-making processes. This lack of transparency can complicate regulatory oversight
and enforcement.

Regulators are increasingly focusing on developing frameworks to address these

challenges, such as implementing stricter rules on trading practices and enhancing
surveillance systems. However, keeping pace with rapidly evolving technologies and
maintaining an effective regulatory environment remains a complex and ongoing task
(Metawa et.al., 2019). Ensuring that computational models and trading algorithms operate
fairly and transparently while protecting market integrity is a critical challenge for both
industry participants and regulators.

In summary, while computational models offer significant advantages in finance, they also
face substantial challenges related to data quality, computational limits, model risk, and
ethical considerations (Lyuu, 2001). Addressing these limitations through rigorous
validation, robust risk management, and effective regulation is essential for harnessing the
full potential of computational models while mitigating associated risks.

7. Future Directions in Computational Finance

The field of computational finance is rapidly evolving, driven by advances in technology
and new methodological approaches. As financial markets become more complex and data-
rich, future developments in computational finance will likely focus on enhancing model
accuracy, computational efficiency, and the ability to adapt to changing market conditions.
Here are some key future directions in the field:

7.1 Emerging Trends and Technologies:

Ǫuantum Computing:
Ǫuantum computing holds promise for revolutionizing computational finance by solving
problems that are currently infeasible for classical computers. Ǫuantum algorithms have
the potential to perform complex financial simulations and optimizations
exponentially faster than traditional methods. For instance, quantum computing could
enhance portfolio optimization by handling more variables and constraints, or improve risk
management models by more accurately simulating extreme market scenarios.
However, practical quantum computers are still in development, and it will take time to
fully realize their potential in finance.

Blockchain Technology:
Blockchain technology offers significant advantages in terms of transparency, security, and
efficiency. In finance, blockchain can streamline processes such as trade
settlement, cross-border transactions, and record-keeping by creating immutable and
transparent ledgers. Smart contracts, which are self-executing contracts with the terms
written into code, can automate and enforce contractual agreements, reducing the need for
intermediaries and lowering transaction costs (Yerashenia & Bolotov, 2019). Blockchain's
potential to improve data integrity and reduce fraud is especially relevant for financial
transactions and regulatory compliance.

7.2 Advancements in Model Accuracy and Computational Power:

High-Performance Computing (HPC):

Advancements in high-performance computing (HPC) enable the processing of larger
datasets and the execution of more complex models in shorter timeframes. HPC systems,
including GPUs and distributed computing platforms, provide the necessary computational
power to run sophisticated financial simulations, machine learning algorithms, and real-
time analytics. Improved computational resources will enhance the ability to analyze vast
amounts of financial data, refine predictive models, and perform more accurate risk
assessments.
Enhanced Machine Learning Techniques:
Machine learning and artificial intelligence continue to evolve, with new techniques
improving model accuracy and adaptability. Advances in deep learning, reinforcement
learning, and natural language processing (NLP) are enhancing the ability to predict market
trends, analyze sentiment, and automate trading strategies. For example, reinforcement
learning algorithms can optimize trading strategies by learning from market interactions,
while NLP techniques can analyze news and social media to gauge market sentiment.
These advancements will lead to more sophisticated and effective financial models.

7.3 The Evolving Role of AI and Machine Learning:

Real-Time Data Analysis:

The ability to process and analyze real-time data is becoming increasingly important in
financial markets. AI and machine learning models are being developed to handle
streaming data and make instantaneous decisions based on live market conditions. This
capability is crucial for high-frequency trading, risk management, and real-time portfolio
adjustments (Neri, 2012). As data sources and technologies improve, real-time analytics
will become more prevalent and impactful in finance.

Explainable AI (XAI):
As AI models become more complex, the need for explainable AI (XAI) is growing. XAI
aims to make AI decision-making processes more transparent and understandable to users.
In finance, this is important for regulatory compliance and gaining trust in AI-
driven models. Techniques for interpretability and explainability will help ensure that AI
models are not only effective but also accountable and transparent.

Integration with Traditional Models:

Future developments will likely focus on integrating AI and machine learning with
traditional financial models to leverage the strengths of both approaches. Hybrid
models that combine statistical methods with machine learning techniques can offer more
comprehensive insights and improve decision-making. For example, integrating machine
learning with econometric models can enhance economic forecasting and risk
management.

In summary, the future of computational finance is marked by exciting advancements in

quantum computing, blockchain technology, and AI. As these technologies mature,
they will enhance model accuracy, computational power, and real-time analytics
capabilities. The integration of emerging trends with traditional financial models will
drive innovation and improve financial decision-making, shaping the next generation of
computational finance.

8. Conclusion

Computational modeling has transformed the finance industry, providing sophisticated

tools for predictive analytics, risk management, and algorithmic trading. While challenges
remain, ongoing advancements in technology and methodology hold
promise for further enhancing the efficacy and reliability of financial models. As the field
evolves, it will be crucial to address ethical and regulatory considerations to
ensure the responsible and equitable use of computational finance.
Computational modeling in finance is a multidisciplinary field that requires knowledge of
finance theories, mathematical models, and computational techniques. By
understanding and leveraging these tools, researchers and practitioners can gain valuable
insights and make informed financial decisions.

9. Consent for publication

All the authors consent to publish this article in the proposed book.

10. Acknowledgement
NA

11. Conflict of Interest

The authors declare that they do not have any conflicts of interest.
9. References
[1] Groesser, S. N., & Jovy, N. (2016). Business model analysis using
computational modeling: a strategy tool for exploration and decision-making.
Journal of Management Control, 27, 61-88.
[2] Cavalcante, R. C., Brasileiro, R. C., Souza, V. L., Nobrega, J. P., & Oliveira,
A. L. (2016). Computational intelligence and financial markets: A survey and
future directions. Expert Systems with Applications, 55, 194-211.
[3] Nolan, J., Parker, D., Van Kooten, G. C., & Berger, T. (2009). An overview of
computational modeling in agricultural and resource economics. Canadian
Journal of Agricultural Economics/Revue canadienne d'agroeconomie, 57(4),
417-429.
[4] Chen, S. H., & Liao, C. C. (2005). Agent-based computational modeling of the
stock price–volume relation. Information Sciences, 170(1), 75-100.
[5] Focardi, S. M., & Fabozzi, F. J. (2004). The mathematics of financial modeling
and investment management (Vol. 138). John Wiley & Sons
[6] Thomaidis, N., & Dash, G. H. (2013). Recent Advances in Computational
Finance (p. 1). Nova Science Publishers, Incorporated.
[7] Axtell, R. L., & Farmer, J. D. (2022). Agent-based modeling in economics and
finance: Past, present, and future. Journal of Economic Literature, 1-101.
[8] Brabazon, A., & O’Neill, M. (2008). Natural computing in computational
finance: An introduction. In Natural Computing in Computational Finance (pp.
1-4). Berlin, Heidelberg: Springer Berlin Heidelberg.
[9] Calder, M., Craig, C., Culley, D., De Cani, R., Donnelly, C. A., Douglas, R., ...
& Wilson, A. (2018). Computational modelling for decision-making: where,
why, what, who and how. Royal Society open science, 5(6), 172096.
[10] Zenios, S. A. (1999). High-performance computing in finance: The last
10 years and the next. Parallel Computing, 25(13-14), 2149-2175.
[11] Bølviken, E. (2014). Computation and modelling in insurance and
finance. Cambridge University Press
[12] Melnik, R. (Ed.). (2015). Mathematical and computational modeling:
With applications in natural and social sciences, engineering, and the arts. John
Wiley & Sons.
[13] Kennedy, D. (2019). Virtual Capital: Computers and the Making of
Modern Finance, 1929- 1975 (Doctoral dissertation, Harvard University).
[14] Hilber, N., Reichmann, O., Schwab, C., & Winter, C. (2013).
Computational methods for quantitative finance: Finite element methods for
derivative pricing. Springer Science & Business Media.
[15] Savine, A., & Andreasen, J. (2021). Modern Computational Finance:
Scripting for Derivatives and xVA. John Wiley & Sons.
[16] Amman, H. M., Kendrick, D. A., Tesfatsion, L., Rust, J., & Judd, K. L.
(Eds.). (1996). Handbook of computational economics: Agent-based
computational economics (Vol. 2). Elsevier.
[17] Binder, A., & Aichinger, M. (2013). A workout in computational
finance. John Wiley & Sons
[18] Chen, S. H. (2005). Computational intelligence in economics and
finance: Carrying on the legacy of Herbert Simon. Information Sciences,
170(1), 121-131.
[19] Lai, T. L., & Xing, H. (2008). Statistical models and methods for
financial markets (Vol. 1017). New York: Springer.
[20] Chang, G. THE EVOLUTION OF THE FINANCE FIELD. In
CONFERENCE PROCEEDINGS BY TRACK (p. 58).
[21] Putriani, D., Ghani, G. M., & Kartiwi, M. (2021). Mark-up vs. Interest-
based financing on gdp: an application of agent-based computational model.
Journal of Islamic Monetary Economics and Finance, 7(1), 55-76.
[22] Tesfatsion, L. (2023). Agent-based computational economics:
Overview and brief history. Artificial intelligence, learning and computation in
economics and finance, 41-58.
[23] Veryzhenko, I. (2012). A reexamination of modern finance issues using
Artificial Market Frameworks (Doctoral dissertation, Université Panthéon-
Sorbonne-Paris I).
[24] Lyuu, Y. D. (2001). Financial engineering and computation: principles,
mathematics, algorithms. Cambridge University Press.
[25] Borodin, A., Mityushina, I., Streltsova, E., Kulikov, A., Yakovenko, I.,
& Namitulina, A. (2021). Mathematical modeling for financial analysis of an
enterprise: Motivating of not open innovation. Journal of Open Innovation:
Technology, Market, and Complexity, 7(1), 79
[26] Agarwal, P. (2019, March). Redefining banking and financial industry
through the application of computational intelligence. In 2019 Advances in
Science and Engineering Technology International Conferences (ASET) (pp.
1-5). IEEE.
[27] Lehtinen, A., & Kuorikoski, J. (2007). Computing the perfect model:
Why do economists shun simulation?. Philosophy of Science, 74(3), 304-329
[28] Metawa, N., Elhoseny, M., Hassanien, A. E., & Hassan, M. K. (2019).
Expert Systems in Finance. New York, NY, USA: Taylor Francis.
[29] Dixon, M., & London, J. (2021). Financial forecasting with α-rnns: a
time series modeling approach. Frontiers in Applied Mathematics and
Statistics, 6, 551138.
[30] Rao, Y., Peng, D., & Tian, T. (2008, October). Advances in Agent-
Based Computational Finance: Aseet Pricing Models from Heterogeneous and
Interactive Perspective. In 2008 4th International Conference on Wireless
Communications, Networking and Mobile Computing (pp. 1-6). IEEE.
[31] Yerashenia, N., & Bolotov, A. (2019, July). Computational modelling
for bankruptcy prediction: Semantic data analysis integrating graph database
and financial ontology. In 2019 IEEE 21st Conference on Business Informatics
(CBI) (Vol. 1, pp. 84-93). IEEE.
[32] Neri, F. (2012). Learning predictive models for financial time series by
using agent based simulations. Transactions on computational collective
intelligence vi, 202-221.