【摘要】 在气候风险加剧背景下，农业保险的合理定价不仅关系到农业生产风险的有效分散，也影响农业保险制度的可持续运行。气候变化是影响作物产量的关键因素，而影响保险费率制定的核心在于对气候波动带来的作物减产率的精准预测。由于不同作物类型和地区对气候变化的敏感性存在差异，作物产量损失预测的精度也存在区域层级不一致性等问题，进而增加了减产率预测和农业保险定价的难度。本文提出以减产率预测为核心的粮食作物农业保险分级费率厘定框架，将不同气候波动条件下的产量预测与减产率估计相结合，并通过多层次预测调和技术解决跨地区产量预测不一致问题。实证研究以冬小麦产量保险为例，基于冬小麦主产区五省产量与气象数据，开展农作物保险气候风险区划及分级费率厘定研究。结果表明，风险分级调整后的费率不仅能更精确地揭示地区间粮食作物产量气候风险的差异，还能显著提升农业保险定价的科学性与针对性，为地区差异化的农业风险管理提供了有力的定量支撑。更多还原

【Abstract】 In the context of increasing climate risks, the rational pricing of agricultural insurance is crucial not only for effectively dispersing agricultural production risks but also for ensuring the sustainable operation of the agricultural insurance system.Climate change is a key factor influencing crop production, and the core of premium rate determination lies with the accurate prediction of yield loss rates resulting from climate variability.However, sensitivities to climate change vary among different crop types and regions, which leads to inconsistencies in the precision of yield loss predictions across spatial hierarchies, thereby increasing the complexity of yield loss estimation and insurance pricing.To address these challenges, this study proposes a multi-level premium rating framework for crop insurance centered on yield loss prediction, which integrates yield forecasting under varying climatic conditions with loss rate estimation and applies forecast reconciliation techniques to resolve cross-regional and hierarchical inconsistencies in yield prediction.Using winter wheat insurance as an empirical example, the study analyzes yield and meteorological data from both provincial and municipal levels across five major winter wheat–producing provinces in northern China to conduct agricultural risk zoning and differentiated premium rate design.The results demonstrate that the proposed premium rates adjusted for risk zoning more accurately captures regional differences in crop yield–climate risk, thereby enhancing the scientific validity and precision of agricultural insurance pricing, and offering effective quantitative support for risk management with regional differentiation.更多还原