BIG DATA AND ECONOMIC FORECASTING: A STATISTICAL APPROACH

Abstract

The rapid expansion of Big Data technologies has fundamentally transformed approaches to economic forecasting and statistical analysis. Traditional forecasting models, which rely on limited and structured datasets, are increasingly insufficient in capturing the complexity, volatility, and non-linearity of modern economic systems. This article examines the role of Big Data in enhancing economic forecasting accuracy through advanced statistical and econometric approaches. Particular attention is paid to the integration of large-scale, high-frequency, and unstructured data sources into statistical models used for macroeconomic and microeconomic predictions. The study analyzes key statistical methods applied in Big Data-driven forecasting, including regression analysis, time-series models, machine learning-based statistical techniques, and predictive analytics. Furthermore, the article evaluates the advantages and limitations of Big Data in economic forecasting, highlighting issues related to data quality, model interpretability, and statistical reliability. The findings suggest that the effective combination of Big Data and statistical methodologies significantly improves forecasting precision, supports evidence-based economic decision-making, and enhances policy formulation in an increasingly data-driven economy.