REGRESSION AND TIME SERIES ANALYSIS OF KAZAKHSTAN’S PRIMARY REAL ESTATE MARKET

Abstract

This study employs regression and time series analysis to study and forecast primary real estate prices in Kazakhstan. By analyzing variables such as average wages, USD/KZT exchange rates, GDP per capita and population growth, the research quantifies their impact on property valuations. The methodology includes rigorous statistical techniques and residual analysis. The Regression model is created by regressing real estate prices on macroeconomic factors non-linearly to capture the relationship adequately. In addition, ARIMA and Cubic Spline regressions are employed to model the trend of housing prices and enhance forecasting accuracy. Comparative analysis indicates that time series models more effectively capture the underlying trend and demonstrate better forecasting abilities. Findings reveal significant correlations between economic indicators and real estate prices, providing a predictive framework for future market trends. This research offers valuable insights for policymakers, investors, and stakeholders in Kazakhstan’s real estate sector, enhancing the understanding of market dynamics through a mathematical lens.