Lab 3
Spatial Regression (GeoDaSpace)
Intro to GeoDaSpace
- Look and feel of the GUI
- Preferences
- Add variables to boxes
- Run basic OLS
Spatial heterogeneity
- No spatial fixed effects in GeoDaSpace
- OLS regime model
Spatial dependence
Spatially lagged exogenous variables
- Create and explore spatial weights
- Create spatially lagged variables
- OLS model with WX
Spatial dependence models
- Spatial lag
- Spatial error
Spatial diagnostics
- OLS
- Moran’s I (check in advanced preferences)
- LM tests
Estimation methods
- OLS
- IV
- ML
- GMM
- VC corrections (White, HAC, KP-Het)
Exercises
- Specify and run a regime model with the counties
- Eplore the differences in the results between including a spatial lag of the
dependent variable and one of the population
References