Residuals dataset: The data set WORK.R may be incomplete

Reply
N/A
Posts: 1

Residuals dataset: The data set WORK.R may be incomplete

Hi All,

I am using version 9.3 of SAS and I am trying to run the code I normally use. This time I am having problems getting the residuals dataset created so that I can test for normality and such. I googled and tried many different things which didn't seem to help. Any ideas would be greatly appreciated. Below is my code and log. Thank you.

CODE:

proc mixed data=WS;

  class Age Country Block Indiv TrmtA TrmtB;

  model ln_16S = Age | TrmtA | TrmtB | Country /ddfm=kr outp=R residual;

  repeated Country Age / subject=Block*TrmtA*TrmtB type=UN@UN;

  lsmeans Country Age Age*TrmtA TrmtB / pdiff adjust=tukey;

  ods listing exclude LSMeans Diffs;

  ods output LSMeans=LSMeans Diffs=Diffs;

run;

title3 'Pairwise comparisons';

%pdmix800(Diffs, LSMeans, sort=yes);

title4 'Residual analysis (symmetry, independence, homoscedasticity)';

proc plot data=R;

  plot StudentResid * (Pred Age TrmtA TrmtB Block Country);

run;

title5 'Residual analysis (normality)';

proc univariate data=R normal;

  var Resid;

  ods select TestsForNormality;

run;

title6 'Residual analysis (outliers)';

proc print data=R;

  where abs(StudentResid) > 3;

run;

LOG

NOTE: With DDFM=SATTERTHWAITE or DDFM=KENWADROGER, unadjusted p-values in tests are based on the

      degrees of freedom specific to that comparison. P-values that are adjusted for multiplicity,

      however, are by default based on the denominator degrees of freedom for the Type 3 test of the

      fixed effect. If you specify the ADJDFE=ROW option in the LSMEANS statement, the adjusted

      p-values take into account the row-wise degrees of freedom.

NOTE: Convergence criteria met.

NOTE: Estimated G matrix is not positive definite.

NOTE: Asymptotic variance matrix of covariance parameter estimates has been found to be singular and

      a generalized inverse was used. Covariance parameters with zero variance do not contribute to

      degrees of freedom computed by DDFM=KENWARDROGER.

ERROR: Cannot write image to ResidualPanel3.png. Please ensure that proper disk permissions are set.

ERROR: Cannot write image to ResidualPanel3.png. Please ensure that proper disk permissions are set.

NOTE: The data set WORK.DIFFS has 3 observations and 16 variables.

NOTE: The data set WORK.LSMEANS has 7 observations and 10 variables.

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

WARNING: The data set WORK.R may be incomplete.  When this step was stopped there were 0 observations

         and 0 variables.

WARNING: Data set WORK.R was not replaced because this step was stopped.

NOTE: PROCEDURE MIXED used (Total process time):

      real time           0.48 seconds

      cpu time 0.20 seconds

2897

2898  title3 'Pairwise comparisons';

2899

2900 %pdmix800(Diffs, LSMeans, sort=yes);

PDMIX800 08.08.2003 processing

  1. 3.0811849556

Tukey-Kramer values for Country  are 0.20303  (avg) 0.20303  (min) 0.20303  (max).

  1. 3.081178758

Tukey  values for Age*TrmtA  are 0.21785  (avg) 0.21785  (min) 0.21785 (max).

  1. 3.081178758

Tukey  values for TrmtB  are 0.21785  (avg) 0.21785  (min) 0.21785 (max).

2901

2902  title3 'Residual analysis (symmetry, independence, homoscedasticity)';

2903

2904  proc plot data=R;

2905 plot StudentResid * (Pred Age TrmtA TrmtB Block Country);

ERROR: Variable STUDENTRESID not found.

ERROR: Variable PRED not found.

ERROR: Variable AGE not found.

ERROR: Variable TRMTA not found.

ERROR: Variable TRMTB not found.

ERROR: Variable BLOCK not found.

ERROR: Variable COUNTRY not found.

NOTE: The previous statement has been deleted.

2906  run;

NOTE: No variables in data set WORK.R.

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

NOTE: PROCEDURE PLOT used (Total process time):

      real time           0.00 seconds

      cpu time 0.00 seconds

2907

2908  title3 'Residual analysis (normality)';

2909

2910  proc univariate data=R normal;

2911 var Resid;

ERROR: Variable RESID not found.

2912 ods select TestsForNormality;

2913  run;

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

NOTE: PROCEDURE UNIVARIATE used (Total process time):

      real time           0.00 seconds

      cpu time 0.00 seconds

WARNING: Output 'TestsForNormality' was not created.  Make sure that the output object name, label, or path is spelled correctly. Also, verify that the appropriate procedure options are used to produce the requested output object.  For example, verify that the NOPRINT option is not used.

2914

2915  title3 'Residual analysis (outliers)';

2916

2917  proc print data=R;

2918 where abs(StudentResid) > 3;

ERROR: Variable StudentResid is not on file WORK.R.

2919  run;

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

NOTE: PROCEDURE PRINT used (Total process time):

      real time           0.00 seconds

      cpu time 0.00 seconds

Respected Advisor
Posts: 2,655

Re: Residuals dataset: The data set WORK.R may be incomplete

I think this all relates to the disk permission error.  Once that hits, everything else just folds up.  Now this may result from the singular covariance matrix, although I can't be sure.  How many parameters are being estimated, and how many records are used?  Is it possible that the convergence is to a local minimum due to overspecification?

At this point, all I can say is that the folks at Tech Services are really, really good and will work with you on this.

Steve Denham

Ask a Question
Discussion stats
  • 1 reply
  • 427 views
  • 1 like
  • 2 in conversation