Pseudo R-squared in GMM lag output

[Julia Koschinsky]

Hello,

 

I have two quick questions: (1) is the pseudo R-squared reported in the GMM lag output of GeoDaSpace a Nagelkerke pseudo R-squared or is it some other form? (2) are there any differences between the pseudo R-squared value and spatial pseudo R-squared in terms of use and interpretation? I haven’t been able to find much detail on the spatial pseudo R-squared value except for a comment indicating that it is calculated using the reduced form of the model to incorporate all feedback effects.

 

Thank you so much!

Jeremy

 

--------

Jeremy Cheung

M.A. Student | School of Criminology

Simon Fraser University | Saywell Hall 10117

8888 University Dr., Burnaby, B.C., V5A 1S6

jeremy...@sfu.ca

 

 

[Pedro Vasconcelos Maia do Amaral]

Dear Jeremy,

 

The pseudo-R^2 in spreg is the squared correlation between the observed and predicted values for the dependent variable.

You can find its formula here:

https://github.com/pysal/spreg/blob/d464bbbc3c8601f1ca1989f4756967dca3a83179/spreg/diagnostics_tsls.py#L129

 

The difference between pseudo-R^2 and spatial pseudo-R^2 is that in the latter the spatial dependence is being filtered out prior to the computation of the R^2. In the standard pseudo-R^2, the actual observed y is considered in the computation of the estimated y. Hence the spatial pseudo-R^2 is preferable. 

 

Best,

 

Pedro.

 

--

Pedro V. Amaral

Cedeplar-FACE-UFMG

‪Av. Antônio Carlos 6627 ‬

‪Belo Horizonte - MG‬

‪31270-901‬

Brazil

--
You received this message because you are subscribed to the Google Groups "Openspace List" group.
To unsubscribe from this group and stop receiving emails from it, send an email to openspace-lis...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/openspace-list/DM6PR11MB322519757A3F93571622115BCD370%40DM6PR11MB3225.namprd11.prod.outlook.com.