proc sql; create table data_want as select * from compare_dataset_214_obsv where wrs_acres not in (select distinct wrs_acres from compare_dataset_148_obs); quit; @Fluvio1 wrote: I tried both programs but they do not provide the results I am looking for.  The attached dataset "Compare_dataset_148_obsv" contains 148 observations. The dataset "Compare_dataset_214_obsv" contains 214 observations.  I would like to identify the observations in  "Compare_dataset_214_obsv" that are not in  "Compare_dataset_148_obsv" and output the observations to a file.  While the datasets contain different variables, they have in common:  WRS_acres.   ... View more

It doesn't make any difference, but if you want the confidence intervals on the original data scale, exponentiate the estimates. That is, if (L(x), U(x)) is the confidence interval for the mean of log(Y) when X=x, then  (exp(L(x)), exp(U(x))) is the confidence interval for the mean of Y. ... View more

Thanks again Rick.  I attached the original paper by Duan that defines the smearing estimator. Also see below:   ... View more