[R-sig-ME] Multilevel with spatial effects

[Linus Holtermann]

Hello,

i found the following working-paper: Corrado/Fingleton "Multilevel Modelling with Spatial Effects" (http://ideas.repec.org/p/str/wpaper/1105.html).
I am interested in fitting a SAR-Multilevel Model as on page 10-11. The inclusion of the group-mean of Y  as exogen variable will result in a endogenity problem. Is it possible to fit these models with existing R-packages? I attach some data for Y, group-mean Y, X and "group". How can i fit the equation (23) on page 10?
Moreover i am interested in fitting a model that is able to identify interactions effects (page 13-14). Is that possible in R?

Another question regards spatial autocorrelation for Level-1 Units. For example, i have districts (Level-1) and regions (Level-2) and i anticipate spatial autocorrelation on Level-1. Can I use a classical spatial weigth matrix (e.g. 8 nearest neigbours (districts) weighted by inverse distance) where neigbours can be districts in the same or nearby regions and incorporate these spatial lag in the multilevel equation? 
In other words, can i replace the Level-1 regression in multilevel model with a classical spatial regression as used in the spdep-packages? Or do i face serious problems like cross-level-endogenity? Would a MCMC approach help?

I know that These are some advanced questions, but i hope someone can help.
Thanks,

best regards,

 
Linus Holtermann
Hamburgisches WeltWirtschaftsInstitut gemeinnützige GmbH (HWWI)
Heimhuder Straße 71
20148 Hamburg
Tel +49-(0)40-340576-336
Fax+49-(0)40-340576-776
Internet: www.hwwi.org
Email: holtermann at hwwi.org
 
AmtsgerichtHamburg HRB 94303
Geschäftsführer: Prof. Dr. Thomas Straubhaar, Gunnar Geyer
Umsatzsteuer-ID: DE 241849425