Hi,
See the attachment of log, from the ODS statement the 3 datasets are generating Cross,Fe and diff10.
The variable CNT has value of 0 and hence 'FE', 'Diff10' are giving warnings.
There is no other option to consider Zeros ,as i already given 'Zeros' in weight statement.
I have no problem with results/Outputs. Is there any option to suppress the warnings in log?
Thanks in Advance.
Srikala
proc freq data=sashelp.class ;
table sex*age/chisq NOWARN; /* CHISQ(WARN=NOLOG) */
run;
Hi Sharp, Thanks for the reply.
But my methods are fisher exact and riskdiff, not Chisq. NOWARN option will not work for me.
IF yours is 2x2 contingency table ,CHISQ will automatically generate FISHER EXACT test.
The warning is there because it is not clear whether a zero entry is a STRUCTURAL zero (eg, the number of men how were pregnant) or a SAMPLING zero (eg, the number of men in the study who had breast cancer). The EXACT statement assumes that it is a sampling zero and considers tables for which that cell is nonzero when computing the p-value.
Available on demand!
Missed SAS Innovate Las Vegas? Watch all the action for free! View the keynotes, general sessions and 22 breakouts on demand.
Learn the difference between classical and Bayesian statistical approaches and see a few PROC examples to perform Bayesian analysis in this video.
Find more tutorials on the SAS Users YouTube channel.