How to calculate pseudo R-Squared in proc GAM?

[issac]

Hi folks;

I've run gam procedure to a data set and got the following output table. I wonder how I can calculate a pseudo R-Squared with the help of deviance.

Number of Observations	49988
Number of Missing Observations	12
Distribution	Gaussian
Link Function	Identity

Final Number of Backfitting Iterations	4
Final Backfitting Criterion	1.6618101E-9
The Deviance of the Final Estimate	240768.87317

  

The backfitting algorithm converged.

Thanks!

Issac

[AllenMcDowell · Accepted solution]

One possibility is discussed in Cameron and Trevedi's "Regression Analysis of Count Data" which describes the deviance as the GLM generalization of the sum of squares. They refer to two papers by Cameron and Windmeijer where a pseudo-R-squared is proposed based on a decomposition of the deviance. In essence, the proposal is to use R2 = 1-(D(intercept-only-model)/D(full-model)). I would recommend reviewing the original papers to verify that the interpretation is appropriate for your particular model before using the result.