论文部分内容阅读
为探讨农作物区域产量保险费率厘定问题,文章利用湖北荆州市1991～2007年县级水稻单产数据,通过建立分层贝叶斯模型,并运用WinBUGS软件进行Gibbs抽样,得出荆州各县市历年水稻单产拟合数据、2008年预测值及其20000个马尔可夫蒙特卡罗(MCMC)模拟值,然后据此厘定各县市水稻区域产量保险费率。结果表明,模型预测能力良好,费率厘定结果有参考价值。
In order to discuss the determination of crop yield insurance premium rates, this paper used county-level rice yield data from 1991 to 2007 in Jingzhou City, Hubei Province. By establishing a hierarchical Bayesian model and using Gibbs sampling with WinBUGS software, Paddy-rice yield fitting data, 2008 forecast value and its 20000 Markov Monte Carlo (MCMC) simulation values, and then determine the rice yield insurance rates in all counties and cities. The results show that the model has good predictive ability and the reference value of rate determination results.