search

pysal / spreg

Spatial econometric regression in Python
https://pysal.org/spreg/
Other
67 stars 23 forks source link

Ambiguity in the definition of Pseudo R-squared and Spatial Pseudo R-squared in spreg.GM_Lag #94

Closed ale-v closed 2 years ago

ale-v commented 2 years ago

The spreg.GM_Lag outputs, among other indicators, a Pseudo R-squared and a Spatial Pseudo R-squared but I cannot understand what they refer to. Is the first the total explained variance attributable to the variables and lagged versions? Is the second the explained variance only attributable to the lagged variables? The official page https://spreg.readthedocs.io/en/latest/generated/spreg.GM_Lag.html does not seem to provide clear information on this. Can anyone please shed some light?

pedrovma commented 2 years ago

Dear @ale-v ,

Please find below excerpts from Anselin & Rey's (2014) book that make clear what [spatial] pseudo-R^2 refers to:

image image image

Best,

Pedro.

ale-v commented 2 years ago

Thank you! It would be probably worth updating the docs though.