The commonly used auto insurance pricing model is the generalized linear model. This model is built on an important assumption that the claims data must be independent.
在车险定价中常用的模型是广义线性模型，该模型建立的一个重要假设是索赔数据的相互独立性。
But in practice, with the policy year increasing, the same policy-holder's claims data in different year has correlation. Therefore the generalized linear model is no longer applicable.
但在实践中，随着投保年份的增加，同一投保人不同年份的索赔数据具有相关性，此时广义线性模型不再适用。
Methods We applied generalized linear mixed models for the nuclear family data to set up the genetic variance component model and estimated parameters using MCMC.
将广义线性混合模型应用于核心家系资料建立遗传方差分量模型，运用MCMC方法进行参数估计。
应用推荐