生育行为的年龄密码OACHSSCD

As China's total fertility rate continues to decline and the population enters a phase of negative growth,it has become increasingly crucial to scientifically forecast the dynamics of fertility levels and improve fertility support policies.Ideally,individuals'intended fertility and their complete fertility should converge.However,discrepancies often exist in reality,making it difficult to accurately predict complete fertility based solely on intended fertility.The age at first birth,as a temporal indicator of an individual's initial fertility behavior,exhibits a strong correlation with both intended fertility and complete fertility.This paper attempts to construct an theoretical framework of"intended fertility-age at first birth-complete fertility".Drawing on three waves of CLDS data(2014,2016,and 2018),we employ the CatBoost machine learning model to capture nonlinear relationships and interaction effects among variables,and further simulate and predict the complete fertility of the childbearing-age population.The study finds that:1)The correlation between age at first birth and complete fertility presents complex nonlinear patterns under the interaction of other socioeconomic variables.2)Incorporating age at first birth into the predictive model can effectively mitigate the bias inherent in estimating complete fertility based solely on intended fertility.3)Particular attention should be paid to the fertility challenges faced by childbearing-age individuals with delayed first births,as well as those residing in urban areas,and in the Northeast and Eastern regions of China.To address the challenge of population decline in China,we propose establishing a"timing-parity"dual-dimensional incentive policy framework.This approach would implement tiered subsidies based on the parity and age at first birth of childbearing-age individuals,and further enhance the targeting and effectiveness of policies by incorporating multiple influencing factors and adapting measures to local conditions.