Research on China insurance demand forecasting: Based on mixed frequency data model

This paper introduces the mixed-frequency data model (MIDAS) to forecast China's insurance demand. A MIDAS regression model is constructed using monthly indicators (Consumer Confidence Index (CCI), China Economic Policy Uncertainty Index (EPU), Consumer Price Index (PPI)) and the quarterly indicator Depth of Insurance (TID). The study investigates the impact and forecasting effects of these indicators on insurance demand, exploring different weighting functions, forecasting windows, and combination methods to optimize accuracy. Results demonstrate the MIDAS model's strong short-term forecasting performance, with rolling window and recursive identification improving accuracy. Consumer confidence is identified as the primary driver of insurance demand, particularly during the COVID-19 period. The study projects China's insurance demand to reach pre-COVID-19 levels by 2023Q2.

Cheng Wang, Mengnan Xu, Zheng Wang, Wenjing Sun