Random Effects for Spatial Panels

[pabloestradac]

This issue discusses the estimation of random effects panel data of spatial models. It proposes two classes: Panel_RE_Lag and Panel_RE_Error to estimate a spatial lag panel model and a spatial error panel model. It follows the estimation procedure detailed in Section 3.3.4 and 3.3.5 from Elhorst - Spatial Econometrics (2014).

The classes Panel_RE_Lag and Panel_RE_Error will be included inside the panel_re.py file. To assess the accuracy of the results, I will compare the estimation with the one obtained in the splm package from R.

The notebook RE_Error_scratch.ipynb contains the estimation for the class Panel_RE_Error.

The notebook RE_Lag_scratch.ipynb contains the estimation for the class Panel_RE_Lag.