Using the proximal policy optimization and prospect theory to train a decision-making model for managing personal finances
The subject of this article is the development of a decision-making model that can, in the future, be incorporated into a personal finance simulator to improve personal finance literacy. The goal of this study is to develop decision-making models tailored to different investor profiles to provide pe...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
National Aerospace University «Kharkiv Aviation Institute»
2024-11-01
|
| Series: | Радіоелектронні і комп'ютерні системи |
| Subjects: | |
| Online Access: | http://nti.khai.edu/ojs/index.php/reks/article/view/2664 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841557704249704448 |
|---|---|
| author | Vladyslav Didkivskyi Dmytro Antoniuk Tetiana Vakaliuk Yevhen Ohinskyi |
| author_facet | Vladyslav Didkivskyi Dmytro Antoniuk Tetiana Vakaliuk Yevhen Ohinskyi |
| author_sort | Vladyslav Didkivskyi |
| collection | DOAJ |
| description | The subject of this article is the development of a decision-making model that can, in the future, be incorporated into a personal finance simulator to improve personal finance literacy. The goal of this study is to develop decision-making models tailored to different investor profiles to provide personalized financial advice on asset allocation. This article employs reinforcement learning techniques and behavioral economics to achieve this objective, thereby contributing to the advancement of practical algorithms and approaches for financial decision-making. The tasks can be formulated as follows: 1) design a reinforcement learning environment featuring different investment options with varying average returns and volatility levels; 2) train the reinforcement learning agent using the Proximal Policy Optimization algorithm to learn recommended investment allocations; 3) implement a reward function based on Prospect Theory, incorporating parameters that reflect different investor risk profiles, such as loss aversion and diminishing sensitivity to gains and losses. The results reveal the development of distinct models for 3 investor profiles: risk-averse, rational, and wealth-maximizing. A graphical analysis of the recommended allocation percentages revealed significant patterns influenced by the value function parameters of Prospect Theory. The practical implications of this research extend to the development of simulation tools based on the model, which will enable individuals to practice and refine their financial strategies in a risk-free environment. These tools bridge the gap in personal finance education by providing experiential learning opportunities. Conclusions. The developed model effectively generates personalized financial advice that reflects individual risk preferences. Future work will focus on creating interactive simulation tools to enhance personal finance management skills. This study underscores the importance of integrating psychological and behavioral insights into financial decision-making models. |
| format | Article |
| id | doaj-art-399aa0e7fe8c419780d617a30c28870f |
| institution | Kabale University |
| issn | 1814-4225 2663-2012 |
| language | English |
| publishDate | 2024-11-01 |
| publisher | National Aerospace University «Kharkiv Aviation Institute» |
| record_format | Article |
| series | Радіоелектронні і комп'ютерні системи |
| spelling | doaj-art-399aa0e7fe8c419780d617a30c28870f2025-01-06T10:47:18ZengNational Aerospace University «Kharkiv Aviation Institute»Радіоелектронні і комп'ютерні системи1814-42252663-20122024-11-012024424825810.32620/reks.2024.4.202369Using the proximal policy optimization and prospect theory to train a decision-making model for managing personal financesVladyslav Didkivskyi0Dmytro Antoniuk1Tetiana Vakaliuk2Yevhen Ohinskyi3Zhytomyr Polytechnic State University, ZhytomyrZhytomyr Polytechnic State University, ZhytomyrZhytomyr Polytechnic State University, ZhytomyrZhytomyr Polytechnic State University, ZhytomyrThe subject of this article is the development of a decision-making model that can, in the future, be incorporated into a personal finance simulator to improve personal finance literacy. The goal of this study is to develop decision-making models tailored to different investor profiles to provide personalized financial advice on asset allocation. This article employs reinforcement learning techniques and behavioral economics to achieve this objective, thereby contributing to the advancement of practical algorithms and approaches for financial decision-making. The tasks can be formulated as follows: 1) design a reinforcement learning environment featuring different investment options with varying average returns and volatility levels; 2) train the reinforcement learning agent using the Proximal Policy Optimization algorithm to learn recommended investment allocations; 3) implement a reward function based on Prospect Theory, incorporating parameters that reflect different investor risk profiles, such as loss aversion and diminishing sensitivity to gains and losses. The results reveal the development of distinct models for 3 investor profiles: risk-averse, rational, and wealth-maximizing. A graphical analysis of the recommended allocation percentages revealed significant patterns influenced by the value function parameters of Prospect Theory. The practical implications of this research extend to the development of simulation tools based on the model, which will enable individuals to practice and refine their financial strategies in a risk-free environment. These tools bridge the gap in personal finance education by providing experiential learning opportunities. Conclusions. The developed model effectively generates personalized financial advice that reflects individual risk preferences. Future work will focus on creating interactive simulation tools to enhance personal finance management skills. This study underscores the importance of integrating psychological and behavioral insights into financial decision-making models.http://nti.khai.edu/ojs/index.php/reks/article/view/2664personal financedecision-makingreinforcement learningprospect theory |
| spellingShingle | Vladyslav Didkivskyi Dmytro Antoniuk Tetiana Vakaliuk Yevhen Ohinskyi Using the proximal policy optimization and prospect theory to train a decision-making model for managing personal finances Радіоелектронні і комп'ютерні системи personal finance decision-making reinforcement learning prospect theory |
| title | Using the proximal policy optimization and prospect theory to train a decision-making model for managing personal finances |
| title_full | Using the proximal policy optimization and prospect theory to train a decision-making model for managing personal finances |
| title_fullStr | Using the proximal policy optimization and prospect theory to train a decision-making model for managing personal finances |
| title_full_unstemmed | Using the proximal policy optimization and prospect theory to train a decision-making model for managing personal finances |
| title_short | Using the proximal policy optimization and prospect theory to train a decision-making model for managing personal finances |
| title_sort | using the proximal policy optimization and prospect theory to train a decision making model for managing personal finances |
| topic | personal finance decision-making reinforcement learning prospect theory |
| url | http://nti.khai.edu/ojs/index.php/reks/article/view/2664 |
| work_keys_str_mv | AT vladyslavdidkivskyi usingtheproximalpolicyoptimizationandprospecttheorytotrainadecisionmakingmodelformanagingpersonalfinances AT dmytroantoniuk usingtheproximalpolicyoptimizationandprospecttheorytotrainadecisionmakingmodelformanagingpersonalfinances AT tetianavakaliuk usingtheproximalpolicyoptimizationandprospecttheorytotrainadecisionmakingmodelformanagingpersonalfinances AT yevhenohinskyi usingtheproximalpolicyoptimizationandprospecttheorytotrainadecisionmakingmodelformanagingpersonalfinances |