Conditional Akaike information criterion in the Fay–Herriot model

摘要: Abstract The Fay–Herriot model, a popular approach in small area estimation, uses relevant covariates to improve the inference for quantities of interest sub-populations. conditional Akaike information (AI) (Vaida and Blanchard, 2005 [23] ) linear mixed-effect models with i.i.d. errors can be extended model measuring prediction performance. In this paper, we derive unbiased AIC (cAIC) three approaches fitting model. cAIC have closed forms are convenient implement. We conduct simulation study demonstrate their accuracy estimating AI superior performance selection than classic AIC. also apply county-level prevalence rates obesity working-age Hispanic females California.