Specifics of predicting the profitability of individual bank products based on machine learning
The article examines the application of machine learning methods for forecasting the profitability of certain banking products. The importance of introducing intelligent algorithms in the banking sector is emphasized, which is due to the need to improve the accuracy of financial forecasts, minimize...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Ukrainian State University of Science and Technologies
2025-06-01
|
| Series: | Економічний вісник Державного вищого навчального закладу Український державний хіміко-технологічний університет |
| Subjects: | |
| Online Access: | http://ek-visnik.dp.ua/wp-content/uploads/pdf/2025-1/Strelchenko.pdf |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | The article examines the application of machine learning methods for forecasting the profitability of certain banking products. The importance of introducing intelligent algorithms in the banking sector is emphasized, which is due to the need to improve the accuracy of financial forecasts, minimize risks and improve strategic management. The relevance of using machine learning in the context of growing competition in the banking sector, tightening regulatory requirements, and the need to strengthen the financial stability of banking institutions is substantiated. It is shown that traditional econometric models have a limited ability to account for nonlinear dependencies, which limits their effectiveness in a rapidly changing economic environment. The study examines the main determinants of the profitability of banking products, including interest margin, operating expenses, credit risk, macroeconomic factors, and regulatory constraints. It explores the use of machine learning to build adaptive predictive models that can identify hidden patterns in financial data and provide more accurate estimates of the future profitability of banking products. The efficiency of key algorithms, including linear and logistic regression, decision trees, ensemble methods and neural networks, is analyzed. It is shown that neural networks have the highest level of predictive accuracy, but their implementation requires significant computational resources. The study proves that the introduction of machine learning in predicting the profitability of banking products helps to improve the accuracy of financial estimates, reduce risks and improve the strategic planning of banking institutions. |
|---|---|
| ISSN: | 2415-3974 2664-2670 |