英文文献：Estimation of crop yield distribution and Insurance Premium using Shrinkage Estimator: A Hierarchical Bayes and Small Area Estimation Approach-利用收缩估计器估计作物产量分布和保险费:一种层次贝叶斯和小面积估计方法
英文文献作者：Awondo, Sebastain Nde,Datta, Gauri S.,Ramirez, Octavio A.,Fonsah, Esendugue Greg
英文文献摘要：
Obtaining reliable estimates of insurance premiums is a critical step in risk sharing and risk transfer necessary to ensure solvency and continuity in crop insurance programs. Challenges encountered in the estimation include dealing with aggregation bias from using county level yield averages as well as properly accounting for spatial and temporal heterogeneity. In this study, we associate some of these challenges as classical small area estimation (SAE) problems. We employ a hierarchical Bayes (HB) SAE to obtain design consistent expected county level yields and Group Risk Plan (GRP) premiums for corm farms in Illinois using quasi-simulated data. Preliminary results show little bias (< 10%) in estimated expected county yields in several counties investigated. We found wide variation in GRP, APH and basis risk across counties for similar level of coverage and scale. Results show that farmers could lower their GRP premiums by as much as 30% by carefully choosing a coverage level and scale combination.

获得可靠的保费估算是风险分担和风险转移的关键步骤，这对确保作物保险计划的偿付能力和连续性是必要的。在估计过程中遇到的挑战包括处理使用县级产量平均值产生的综合偏差，以及适当地考虑空间和时间异质性。在本研究中，我们将这些挑战与经典的小面积估计(SAE)问题联系起来。利用准模拟数据，我们采用分层Bayes (HB) SAE来获得伊利诺斯州球茎农场设计一致的预期县级产量和群体风险计划(GRP)溢价。初步结果显示，在调查的几个县，估计的预期县产量几乎没有偏差(< 10%)。我们发现，在相同的覆盖率和规模下，跨县的GRP、APH和基础风险存在很大差异。结果表明，通过谨慎选择覆盖水平和规模组合，农民可以降低高达30%的GRP保费。