Search

The article provides information on the model selection for geostatistical models. Though ecologists and scientists consider a number of plausible models in statistical applications, formal consideration of model selection in ecological applications has increased in the recent years. The researchers have identified two problems of spatial modeling and model selection. They used a geostatistical model to predict a response at unobserved locations. This method allows any number of the explanatory variables observed at the sample locations to be included in the model to improve the entire predictions. The impact of ignoring spatial correlation on the selection of explanatory variables are considered.