A Predictive Data Mining Framework for Currency Crisis Management

Currency and financial crises, as fundamental challenges in macroeconomics, can have far-reaching consequences for the economic and social stability of societies. Identifying and predicting these crises with the help of predictive models can assist policymakers in making preventive decisions and crisis management. This research aims to design and analyze predictive models for identifying currency crises. In this regard, data related to key macroeconomic variables such as exchange rates, stock market indices, and bank interest rates have been collected and preprocessed. Then, machine learning algorithms such as decision trees have been employed for modeling. The results show that the proposed model can accurately predict financial crises and identify the relative importance of economic variables. Sensitivity analysis of the model reveals that exchange rate fluctuations have a significant impact on crisis detection. Finally, suggestions are provided to strengthen information infrastructure, utilize more advanced algorithms, and integrate qualitative and quantitative methods to improve the prediction and management of economic crises. This research provides a practical model for policymakers and economic researchers.

کلیدواژه ها:

Predictive Data Mining ، Financial Crisis Management ، Currency Crisis ، Crisis Prediction ، Machine Learning ، Economic Modeling