How to fix ERROR: Error in estimation routine on SAS Studio OnDemand for Academics

[Joa14]

I am using the Tasks Option on SAS Studio On Demand and choosing Linear Models → Generalized Linear Models and running the model with Distribution:  Zero Inflated Negative Binomial. The problem is that the Wald Chi-Square and the Pr > ChiSq results are blank. I expect these rows to have values. Some variables ran successfully on their own but fail when combined with others. Additionally, models that previously ran without issues are now failing generating the same error.  The error message is: ERROR: Error in estimation routine (See the Log and Screenshot of results attached). All help would be greatly appreciated. Thanks!

The Log is:

ods noproctitle;
ods graphics / imagemap=on;

proc genmod data=VEGCH2.'BPSD50PLOTS_RA&BA_SHORT'n plots=(predicted 
		resraw(index) );
	model BPSdCnt=EricShrRA CO / dist=zinb;
	zeromodel EricShrRA CO;
run;

1          OPTIONS NONOTES NOSTIMER NOSOURCE NOSYNTAXCHECK;

NOTE: ODS statements in the SAS Studio environment may disable some output features.

73        

74         /*

75          *

76          * Task code generated by SAS Studio 3.8

77          *

78          * Generated on '8/1/24, 12:31 PM'

79          * Generated by 'u42887087'

80          * Generated on server 'ODAWS02-USW2-2.ODA.SAS.COM'

81          * Generated on SAS platform 'Linux LIN X64 5.14.0-284.30.1.el9_2.x86_64'

82          * Generated on SAS version '9.04.01M7P08062020'

83          * Generated on browser 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko)

83       ! Chrome/127.0.0.0 Safari/537.36'

84          * Generated on web client

84       ! 'https://odamid-usw2-2.oda.sas.com/SASStudio/main?locale=en&zone=GMT-04%253A00&ticket=ST-149514-Jta4f...'

85          *

86          */

87        

88         ods noproctitle;

89         ods graphics / imagemap=on;

90        

91         proc genmod data=VEGCH2.'BPSD50PLOTS_RA&BA_SHORT'n plots=(predicted

92         resraw(index) );

93         model BPSdCnt=EricShrRA CO / dist=zinb;

94         zeromodel EricShrRA CO;

95         run;

WARNING: The specified model did not converge.

ERROR: Error in estimation routine.

NOTE: The negative binomial dispersion parameter was estimated by maximum likelihood.

ERROR: Domain error.

ERROR: Termination due to Floating Point Exception

NOTE: The SAS System stopped processing this step because of errors.

NOTE: PROCEDURE GENMOD used (Total process time):

      real time           0.07 seconds

      user cpu time       0.06 seconds

      system cpu time     0.00 seconds

      memory              5388.62k

      OS Memory           30636.00k

      Timestamp           08/01/2024 04:31:31 PM

      Step Count                        29  Switch Count  0

      Page Faults                       0

      Page Reclaims                     1600

      Page Swaps                        0

      Voluntary Context Switches        8

      Involuntary Context Switches      2

      Block Input Operations            0

      Block Output Operations           488

96        

97         OPTIONS NONOTES NOSTIMER NOSOURCE NOSYNTAXCHECK;

109        

[Joa14]

I am using the Tasks Option on SAS Studio On Demand and choosing Linear Models → Generalized Linear Models and running the model with Distribution:  Zero Inflated Negative Binomial. The problem is that the Wald Chi-Square and the Pr > ChiSq results are blank. I expect these rows to have values. Some variables ran successfully on their own but fail when combined with others. Additionally, models that previously ran without issues are now failing generating the same error.  The error message is: ERROR: Error in estimation routine (See the Log and Screenshot of results attached). All help would be greatly appreciated. Thanks!

The Log is:

1          OPTIONS NONOTES NOSTIMER NOSOURCE NOSYNTAXCHECK;

ods noproctitle;
ods graphics / imagemap=on;

proc genmod data=VEGCH2.'BPSD50PLOTS_RA&BA_SHORT'n plots=(predicted 
		resraw(index) );
	model BPSdCnt=EricShrRA CO / dist=zinb;
	zeromodel EricShrRA CO;
run;

NOTE: ODS statements in the SAS Studio environment may disable some output features.

73        

74         /*

75          *

76          * Task code generated by SAS Studio 3.8

77          *

78          * Generated on '8/1/24, 12:31 PM'

79          * Generated by 'u42887087'

80          * Generated on server 'ODAWS02-USW2-2.ODA.SAS.COM'

81          * Generated on SAS platform 'Linux LIN X64 5.14.0-284.30.1.el9_2.x86_64'

82          * Generated on SAS version '9.04.01M7P08062020'

83          * Generated on browser 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko)

83       ! Chrome/127.0.0.0 Safari/537.36'

84          * Generated on web client

84       ! 'https://odamid-usw2-2.oda.sas.com/SASStudio/main?locale=en&zone=GMT-04%253A00&ticket=ST-149514-Jta4f...'

85          *

86          */

87        

88         ods noproctitle;

89         ods graphics / imagemap=on;

90        

91         proc genmod data=VEGCH2.'BPSD50PLOTS_RA&BA_SHORT'n plots=(predicted

92         resraw(index) );

93         model BPSdCnt=EricShrRA CO / dist=zinb;

94         zeromodel EricShrRA CO;

95         run;

WARNING: The specified model did not converge.

ERROR: Error in estimation routine.

NOTE: The negative binomial dispersion parameter was estimated by maximum likelihood.

ERROR: Domain error.

ERROR: Termination due to Floating Point Exception

NOTE: The SAS System stopped processing this step because of errors.

NOTE: PROCEDURE GENMOD used (Total process time):

      real time           0.07 seconds

      user cpu time       0.06 seconds

      system cpu time     0.00 seconds

      memory              5388.62k

      OS Memory           30636.00k

      Timestamp           08/01/2024 04:31:31 PM

      Step Count                        29  Switch Count  0

      Page Faults                       0

      Page Reclaims                     1600

      Page Swaps                        0

      Voluntary Context Switches        8

      Involuntary Context Switches      2

      Block Input Operations            0

      Block Output Operations           488

96        

97         OPTIONS NONOTES NOSTIMER NOSOURCE NOSYNTAXCHECK;

109        

[Kurt_Bremser]

Please do not double-post.

Have you checked this:

WARNING: The specified model did not converge.

This may cause such procedures to run into infinite loops.

[Joa14]

I have not. I have seen another warning when running the models: WARNING: Negative of Hessian not positive definite.

[Ksharp]

There are too many reason to produce this NOTE.
Check this paper at page 10.
NOTE: Convergence criteria met but final hessian is not positive definite.


https://support.sas.com/resources/papers/proceedings12/332-2012.pdf

[Joa14]

I am using the Tasks Option on SAS Studio On Demand and choosing Linear Models → Generalized Linear Models and running the model with Distribution:  Zero Inflated Negative Binomial. The problem is that the Wald Chi-Square and the Pr > ChiSq results are blank. I expect these rows to have values. Some variables ran successfully on their own but fail when combined with others. Additionally, models that previously ran without issues are now failing generating the same error.  The error message is: ERROR: Error in estimation routine (See the Log and Screenshot of results attached). All help would be greatly appreciated. Thanks!

ods noproctitle;
ods graphics / imagemap=on;

proc genmod data=VEGCH2.'BPSD50PLOTS_RA&BA_SHORT'n plots=(predicted 
		resraw(index) );
	model BPSdCnt=EricShrRA CO / dist=zinb;
	zeromodel EricShrRA CO;
run;

 The Log is:

1          OPTIONS NONOTES NOSTIMER NOSOURCE NOSYNTAXCHECK;

NOTE: ODS statements in the SAS Studio environment may disable some output features.

73        

74         /*

75          *

76          * Task code generated by SAS Studio 3.8

77          *

78          * Generated on '8/1/24, 12:31 PM'

79          * Generated by 'u42887087'

80          * Generated on server 'ODAWS02-USW2-2.ODA.SAS.COM'

81          * Generated on SAS platform 'Linux LIN X64 5.14.0-284.30.1.el9_2.x86_64'

82          * Generated on SAS version '9.04.01M7P08062020'

83          * Generated on browser 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko)

83       ! Chrome/127.0.0.0 Safari/537.36'

84          * Generated on web client

84       ! 'https://odamid-usw2-2.oda.sas.com/SASStudio/main?locale=en&zone=GMT-04%253A00&ticket=ST-149514-Jta4f...'

85          *

86          */

87        

88         ods noproctitle;

89         ods graphics / imagemap=on;

90        

91         proc genmod data=VEGCH2.'BPSD50PLOTS_RA&BA_SHORT'n plots=(predicted

92         resraw(index) );

93         model BPSdCnt=EricShrRA CO / dist=zinb;

94         zeromodel EricShrRA CO;

95         run;

WARNING: The specified model did not converge.

ERROR: Error in estimation routine.

NOTE: The negative binomial dispersion parameter was estimated by maximum likelihood.

ERROR: Domain error.

ERROR: Termination due to Floating Point Exception

NOTE: The SAS System stopped processing this step because of errors.

NOTE: PROCEDURE GENMOD used (Total process time):

      real time           0.07 seconds

      user cpu time       0.06 seconds

      system cpu time     0.00 seconds

      memory              5388.62k

      OS Memory           30636.00k

      Timestamp           08/01/2024 04:31:31 PM

      Step Count                        29  Switch Count  0

      Page Faults                       0

      Page Reclaims                     1600

      Page Swaps                        0

      Voluntary Context Switches        8

      Involuntary Context Switches      2

      Block Input Operations            0

      Block Output Operations           488

96        

97         OPTIONS NONOTES NOSTIMER NOSOURCE NOSYNTAXCHECK;

109