03-27-2017 06:00 PM - edited 03-27-2017 06:14 PM
I have the following SAS statement that yields a 0 for std error of mean and sum. I am trying to understand why. Any help would be appreciated. I was able to get output if I set varmethod to jackknife. Does that help?
data sales_str_sizes; input str_sales _total_; datalines; 1 425 2 20 3 6 4 2 5 2 6 2 7 43 ; proc surveymeans data=sales_neyman mean sum clsum clm total=sales_str_sizes; var sales; weight SamplingWeight; strata str_sales; title "Neyman Allocation for Sales"; run;
03-28-2017 05:44 PM
The answer probably lies in the data summarized.
Note that part of the reason to use a TOTAL secondary set is to generate a finite population correction but this is not done if you ue BRR or jacknife. So that would be part of the reason you get a different result with jackknife: no correction.
Having strata with a population of 2 such as your strata 4, 5 and 6 looks a tad odd to me.
Need further help from the community? Please ask a new question.