Bayesian Spatial Analysis of Misclassified Binary Data Incorporating Internal‐Validation Studies

Accurate spatial classification is a challenging task, especially when binary outcomes are subject to measurement errors and misclassification. Motivated by a precipitation study in South Korea, we propose Bayesian spatial classification methods with misclassification correction using internal validation data. The prior distributions for the misclassification parameters are specified using internal validation data in the Bayesian spatial classification of the main study, where the gold‐standard device is unavailable. A simulation study is conducted to compare the performance of the proposed methods with the naive method that ignores the misclassification. It is found that the proposed methods outperform the naïve model. The proposed methods are also illustrated with precipitation data from South Korea.