Bayesian Model Averaging with the Integrated Nested Laplace Approximation

Round 1

Reviewer 1 Report

see attached Acrobat PDF

Comments for author File: Comments.pdf

Author Response

Please, see attached file with the reply to the editor and the reviewers.

Author Response File: Author Response.pdf

Reviewer 2 Report

See enclosed file

Comments for author File: Comments.pdf

Author Response

Please, see attached file with the reply to the editor and the reviewers.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

I think the authors did a good job of responding to my initial review comments.

Author Response

Many thanks for agreeing with our changes.

Reviewer 2 Report

See attached file

Comments for author File: Comments.pdf

Author Response

We have now followed the reviewer’s comment and describe the full model introduced on page 2 as the generalized nesting spatial model (GNS) following Halleck Vega, Solmaria and J. Paul Elhorst (2015). Then, we have introduced the SAC model as the GNS model after dropping the term on the lagged covariates. On page 4, we have referred to the GNS when introducing the linear predictor of the model. We believe that this is still accurate and that the reader will understand at this point that the SAC model will have the lagged covariates term removed in the linear predictor. This will make the paper address a more general problem but, at the same time, make it clear that we are focusing on the SAC model for the examples.

Furthermore, we have also fixed the typos mentioned by this referee about the period missing after some equations.

References:

Halleck Vega, Solmaria and J. Paul Elhorst. 2015. The SLX model. Journal of Regional Science 55(3), 339–363.

doi:10.1111/jors.12188.