Hi,

I am running a hedonic regression on how (log) house prices and distance from a local airport. Because of the spatial relationship between houses I need to use a SAR model.

My data is cross-sectional and contains the latitude and longitude of several homes (I do not have a shapfile).

I am trying to create an inverse-distance matrix but am met with the error "Two or more observations have the same coordinates" which makes sense as some houses share a postcode hence have the same latitude and longitude.

How do I go about creating this matrix? I read about dropping duplicates, creating the matrix and then merging the data sets together but this causes me to lose observations?

Any help would be greatly appreciated!