Comparative Analysis of Machine Learning Algorithms for Inflation Rate Classification and Economic Trend Forecasting

Predictive modeling of inflation rates is critical for economic policy and risk management. This study compares the efficacy of K-Nearest Neighbors (KNN) and logistic regression algorithms in classifying inflation into 'Low,' 'Medium,' and 'High' categories, which are essential for monitoring and managing economic stability. Through meticulous preprocessing, including outlier removal, normalization, and imputation, the dataset was refined for accurate analysis. The study established inflation thresholds using empirical data and economic theory to create structured categories. The EDA revealed the fluctuating nature of inflation and its long-term trends, providing a clear picture of economic conditions. Logistic regression demonstrated a higher accuracy rate and better performance metrics over KNN, suggesting it as a superior model for inflation classification. The findings emphasize the importance of precise model selection in economic forecasting and propose logistic regression as a robust tool for policy-makers and economists in strategic economic planning.

Перевод пока недоступен