PROC GENMOD

[TANMOY05]

I am getting 

ERROR: Error in computing the variance function.

ERROR: Error in parameter estimate covariance computation.

ERROR: Error in estimation routine.

in my log  ? how to solve this?

ods output lsmeans=lsm_chg Diffs=diff ESTIMATE=diff_chg ;

326 proc genmod data=lb3tmp DESCENDING;

327 by stratum;

328 class cohort visitn subject ;

329 model resp(EVENT='1') = visitn cohort cohort*visitn base / LINK= logit DIST=BIN ;

330 repeated subject=subject /type=UN ;

331 lsmeans cohort / DIFF cl ;

332

333 run ;

NOTE: PROC GENMOD is modeling the probability that resp='1'.

NOTE: Algorithm converged.

ERROR: Error in computing the variance function.

ERROR: Error in parameter estimate covariance computation.

ERROR: Error in estimation routine.

NOTE: The above message was for the following BY group:

STRATUM=CKD

WARNING: Output 'ESTIMATE' 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.

WARNING: Output 'Diffs' 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.

WARNING: Output 'lsmeans' 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.

NOTE: PROCEDURE GENMOD used (Total process time):

real time 0.82 seconds

cpu time 0.80 seconds

The below mentioned pic is the lb3tmp dataset.

[PaigeMiller]

Your input data set has only 6 observations?

[TANMOY05]

No, it continues till 100 rows, though these 6 observations repeats itself with change in values for different subjects and the resp column is contains values 0 and 1 only.

[ballardw]

You don't show how many records have Stratu=CKD. One suspects that you don't have enough records in the stratum to complete the calculations needed or perhaps no Event=1 for the stratum.

Look at your data for stratum CKD very closely.

[TANMOY05]

By enough , u mean by how much 1 should be there to be able to show results?