在联合广义线性模型中,散度参数与均值都被赋予了广义线性模型的结构,本文主要考虑在只有分布的一阶矩和二阶矩指定的条件下,联合广义线性模型中均值部分的变量选择问题。本文采用广义拟似然函数,提出了新的模型选择准则(EAIC);该准则是Akaike信息准则的推广。论文通过模拟研究验证了该准则的效果。
This paper focuses on variable selection and proposes a new model selection criterion for joint generalized linear models with structured dispersions. The proposal is based on the extended quasi-likelihood that only needs to specify first two moments of the distribution. The new criterion (EAIC) is an extension of Akaike' s information criterion (AIC). Its performance is investigated through simulation studies, and the results support the utility of the methodology.