topic Re: "None of the EDF goodness-of-fit statistics for the Gamma distribution are available." in SAS Programming

"None of the EDF goodness-of-fit statistics for the Gamma distribution are available." note.

Dmitry — Wed, 09 Aug 2017 00:05:42 GMT

I am trying to fit a set of observed values into Gamma distribution, using UNIVARIATE:

proc univariate data=calls;
	var iat;
	histogram  / gamma;
run;

but getting no goodness-of-fit tests, just the note in the subject. Other outputs (basic measures, fitted gamma parameters, quantiles) are produced.

But when I try th example code from here:

http://support.sas.com/documentation...

data Plates;
   label Gap = 'Plate Gap in cm';
   input Gap @@;
   datalines;
0.746  0.357  0.376  0.327  0.485 1.741  0.241  0.777  0.768  0.409
0.252  0.512  0.534  1.656  0.742 0.378  0.714  1.121  0.597  0.231
0.541  0.805  0.682  0.418  0.506 0.501  0.247  0.922  0.880  0.344
0.519  1.302  0.275  0.601  0.388 0.450  0.845  0.319  0.486  0.529
1.547  0.690  0.676  0.314  0.736 0.643  0.483  0.352  0.636  1.080
;
run;

ods select ParameterEstimates GoodnessOfFit FitQuantiles MyHist;
proc univariate data=Plates;
   var Gap;
   histogram / midpoints=0.2 to 1.8 by 0.2
               lognormal
               weibull
               gamma
               vaxis   = axis1
               name    = 'MyHist';
   inset n mean(5.3) std='Std Dev'(5.3) skewness(5.3)
          / pos = ne  header = 'Summary Statistics';
   axis1 label=(a=90 r=0);
run;

it works - all three GOF metrics (KS, Cramer, Anderson-Darling) are produced.

So, is it something in my data that does not allow to compute these metrics?

Re: "None of the EDF goodness-of-fit statistics for the Gamma distribution are available."

Ksharp — Wed, 09 Aug 2017 13:20:59 GMT

Maybe your variable iat have some negative value which could not be gamma distribution.

or try PROC GENMOD .

data Plates;
   label Gap = 'Plate Gap in cm';
   input Gap @@;
   datalines;
0.746  0.357  0.376  0.327  0.485 1.741  0.241  0.777  0.768  0.409
0.252  0.512  0.534  1.656  0.742 0.378  0.714  1.121  0.597  0.231
0.541  0.805  0.682  0.418  0.506 0.501  0.247  0.922  0.880  0.344
0.519  1.302  0.275  0.601  0.388 0.450  0.845  0.319  0.486  0.529
1.547  0.690  0.676  0.314  0.736 0.643  0.483  0.352  0.636  1.080
;
run;

proc genmod data=Plates;
   model Gap= /dist=gamma;
run;

Re: "None of the EDF goodness-of-fit statistics for the Gamma distribution are available."

Rick_SAS — Wed, 09 Aug 2017 14:21:38 GMT

Yes, most likely your data is causing the issue. The table "Availability of EDF Tests" in the UNIVARIATE documentation shows the situations for which the EDF tests are available. In your example, theta=0 (known) and the shape and scale parameters are unknown. The table shows that EDF tests are only available when the estimate for the shape parameter is greater than 1.

I am going to guess that your shape estimate (alpha) is less than 1.