Improving upon the effective sample size based on Godambe information for block likelihood inference

We consider the effective sample size, based on Godambe information, for block likelihood inference which is an attractive and computationally feasible alternative to full large correlated datasets. With reference a Gaussian random field having constant mean, we explore how choice of blocks impacts this size. This done by introducing column-wise blocking method spreads out spatial points within each block, instead keeping them close together as existing row-wise does. It seen that can lead considerable gains in size efficiency compared blocking, while retaining computational simplicity. Analytical results direction are obtained under AR (1) model. The insights so found facilitate study other one-dimensional correlation models well plane, where closed form expressions intractable. Simulations provide support our conclusions.