[R-sig-Geo] R: spml function and the spatial.error option

[Ariel Ortiz-Bobea]

Hello Giovanni,

I finally found what was wrong and (as expected) it had nothing to do with
the function.

I turns out that the reported S.E.s converged to the real ones when I used a
made up sp object or the US county map from the "map" package. 

On the contrary, when I started from a US shape file the reported S.E.s did
not match the ones from the Monte Carlo simulation. My guess is that
somewhere down the line the sp object I had created from the shape file was
passing incorrect information to the spml function. I did not dig down to
the root but it is certainly having to do with the spatial object and how I
constructed the neighbors and weights from it.

Anyways, thanks for the help and I hope this helps other people in the
future.

Cheers,

Ariel

-----
Ariel Ortiz-Bobea
PhD Candidate in Agricultural & Resource Economics
University of Maryland - College Park
--
View this message in context: http://r-sig-geo.2731867.n2.nabble.com/R-spml-function-and-the-spatial-error-option-tp7526451p7579082.html
Sent from the R-sig-geo mailing list archive at Nabble.com.