Hao Tian, Xia Li
As China's aging population accelerates (with those aged 60 and above reaching 310 million by the end of 2024, accounting for 22% of the total population), the silver economy has emerged as a key driver of market transformation. However, the silver consumer market faces “supply mismatch” issues such as inadequate adaptation of smart products and homogenized services.This study employed stratified cross-subsample sampling to ensure sample representativeness. After data preprocessing and model parameter optimization, a random forest regression model was applied to 2,140 valid survey responses from seniors (aged 60+) across six core urban districts in the Beijing-Tianjin-Hebei, Yangtze River Delta, and Pearl River Delta megaregions. (Test set R² = 0.79, MAE = 45.8 yuan). The model systematically analyzed monthly consumption determinants among seniors and constructed a predictive framework. Results indicate that income level, age cohort, and technology acceptance are core determinants. High-income seniors exhibit significantly higher average monthly spending than low-income counterparts. MAE=45.8 yuan), systematically analyzing monthly consumption factors and constructing a predictive model. Results indicate that income level, age group, and technology acceptance are core determinants. High-income seniors spend 2.2 times more monthly than low-income seniors, while those aged 60-65 consume 63% more than those over 76. and those with a technology acceptance score of 5 spent 42% more than those with a score of 1. This study addresses the existing literature gap in capturing the nonlinear relationships within silver-haired consumption, providing empirical support for targeted policy implementation and enterprise-level product design for the elderly, thereby contributing to the high-quality development of the silver economy.Keywords: Silver economy; Random forest model; Factors influencing consumption; Group heterogeneity; Three major metropolitan areas; Monthly consumption prediction
Silver Economy; Random Forest Model; Factors Influencing Consumption; Three Major Urban Agglomerations; Monthly Consumption Forecast