ANALYSIS OF ALGORITHMIC AND MATHEMATICAL APPARATUS FOR SYSTEM OF DEVELOPMENT AND ANALYSIS OF PERSONAL FINANCE MANAGEMENT TOOLS
DOI:
https://doi.org/10.32689/maup.it.2022.3.4Keywords:
Personal finance, tools personal finance management, reinforcement learning, linear regression, decision trees, polynomial regressionAbstract
Personal finance management is important for people nowadays. It determines financial decision making so that a person feels protected, improves personal financial well-being, and creates a reserve fund in advance to handle unforeseen situations. Various financial instruments can be used to achieve financial goals. It can be a bank deposit, bonds, shares, real estate, etc. Each financial instrument also has its own characteristics. Each instrument has certain riskiness, time limits and profitability. Currency risks, financial crises, and pandemics greatly affect the characteristics of financial instruments, make them more or less attractive. Investing in government bonds can be very profitable in one time period and less profitable in other time period. So, financial instruments may be suitable for certain financial goals in one time period and not suitable in another. The question arises of choosing a mathematical apparatus for solving the problem of finding optimal strategies for people’s financial investments, in order to consider various factors. Such methods as Supervised learning, Unsupervised learning, and Reinforcement learning and their peculiarities are considered in detail. It is also considered certain Supervised learning algorithms: linear regression, decision trees, polynomial regression. Algorithms of the Unsupervised learning method: cluster analysis, K-means method, hierarchical clustering, densitybased clustering. The approach of agent modeling and the method of statistical modeling of random processes using Markov chains, reinforcement learning algorithms, namely the Value-based method and the Policy-based method, are also considered. Different types of problems that can be solved using the above approaches were also considered. These algorithms can be widely used in various areas, as well as for solving personal finance management tasks.
References
R. S. Sutton, and A. G. Barto, Reinforcement learning: an introduction. Cambridge: The MIT Press, 2015, p. 143-160.
Sutton R. S., & Barto A. G. Introduction to reinforcement learning (Vol. 135). Cambridge: MIT press; 1998.
Ronald van Loon. Machіne learnіng explained: Understanding supervised, unsupervіsed, and reinforcement learning. URL: https://bіgdata-madesіmple.com/machine-learning-explaіned-understanding-supervised-unsupervisedand-reinforcement-learnіng.
Антонюк Д. С., Вакалюк Т. А., Дідківський В. В., Візгалов О. Ю., Необхідність розробки симулятора управління персональними фінансами. Інноваційна педагогіка: науковий журнал. Вип. 24. Том 2. Видавничий дім «Гельветика», 2020. С. 208-212.
Кононова К. Ю. Машинне навчання: методи та моделі. МОН України, Харківський національний університет імені В. Н. Каразіна, 2020. С. 47-48
What are the types of Reinforcement learning algorithms? URL: https://www.finsliqblog.com/ai-and-machinelearning/what-are-the-types-of-reinforcement-learning-algorithms/
B. Hambly, R. Xu, H. Yang, Recent Advances in Reinforcement Learning in Finance, 2021 URL: https://arxiv.org/abs/2112.04553
Unsupervised learning URL: https://en.wikipedia.org/wiki/Unsupervised_learning.
Reinforcement learning URL: https://en.wikipedia.org/wiki/Reinforcement_learning.
Markov decision process URL: https://en.wikipedia.org/wiki/Markov_decision_process.
Q-learning URL: https://en.wikipedia.org/wiki/Q-learning.
Value-based Methods in Deep Reinforcement Learning. URL: https://towardsdatascience.com/value-basedmethods-in-deep-reinforcement-learning-d40ca1086e1
Machine Learning and Linear Models: How They Work. URL: https://blog.dataiku.com/top-machine-learningalgorithms-how-they-work-in-plain-english-1
Introduction to Markov Chains. URL: https://towardsdatascience.com/introduction-to-markov-chains-50da3645a50d
Introducing the AI Economist: Why Salesforce Researchers are Applying Machine Learning to Economics. URL: https://www.salesforce.com/news/stories/introducing-the-ai-economist-why-salesforce-researchers-are-applyingmachine-learning-to-economics.
