Forecasting Industrial Growth Using Multi-Stage Regression and Elasticity Analysis

General Background: Forecasting the development of industrial enterprises is crucial for strategic planning and sustainable growth. Specific Background: Existing models employ mathematical statistics, economic-mathematical modeling, and regression analysis to predict key performance indicators. Knowledge Gap: However, there remains limited integration of multi-stage regression methods and elasticity analysis in forecasting industrial output over long-term periods. Aims: This study aims to construct a multifactorial forecasting model using multiple regression and correlation functions to estimate the gross product volume of industrial enterprises in Andijan region from 2024 to 2030. Results: The model, based on historical data from 2010–2023, achieved a high coefficient of determination (R = 0.962) and an acceptable forecast error (2–8%). Elasticity coefficients indicate consistent growth in production efficiency, despite fluctuations in capital fund efficiency. Novelty: The use of multi-stage regression and elasticity-based adjustments, combined with statistical extrapolation, offers a more accurate and regionally contextualized forecast of industrial performance. Implications: The findings support targeted strategic planning by highlighting district-level disparities and recommending policy measures such as resource reallocation, entrepreneurship development, and production modernization to ensure sustainable industrial growth through 2030. Highlights: High accuracy forecast using regression with low error range. Elasticity analysis improves long-term industrial predictions. Informs strategic planning for regional enterprise growth. Keywords: industrial forecasting, multiple regression, elasticity coefficient, production modeling, economic extrapolation