Interpreting marginal effects output in Spatial Durbin Model

[Craig Carpenter]

Hi there,

I am using the Stata command spregsdm to implement the Spatial Durbin Model to calculate the average direct and average indirect effects. Per LeSage and Pace (2009), I thought I understood how to interpret the results, but I don't understand why the results include direct and indirect effects for the spatial lags of the explanatory variables (and direct and indirect effects for the non-lagged explanatory variables). I thought that the "indirect effects" represent the effects of the spatial lags.

Any help would be appreciated.

Thanks,
Craig