Forecasting China's Real Estate Investment Trends: A Dynamic Regression Model Approach with ARIMA

Abstract

This investigation utilizes a dynamic regression model with Autoregressive Integrated Moving Average (ARIMA) error correction to evaluate and forecast the trajectory of China's real estate market. The model interprets investment cycles, indicating an imminent significant peak and anticipates subsequent market adjustments. The research integrates key indicators such as production output and sales data to substantiate its predictions, simultaneously warning of an overinflated housing market due to shifting demands and demographic evolution. The authors argue for a more sophisticated modeling approach to accurately reflect the market's nuances and propose targeted policy measures aimed at maintaining equilibrium. The study presents the model's formidable forecasting ability while also acknowledging the complex, multifaceted nature of the real estate sector which mandates ongoing model enhancements. This dual focus on predictive accuracy and the call for adaptive modeling underscores the study's commitment to providing actionable insights in a rapidly changing economic landscape. The proposed policies seek to preemptively counteract instability, thereby contributing to a more sustainable real estate economy in China.