Hi Minze,

What method or code do you use to kick those 9 out? You might use almost the same code to present their ID's. Can you possibly write them to a different dataset?

If all that is impossible you would have to compare the input and output datasets using PROC COMPARE.

Regards - Jim. -- Jim Groeneveld, Netherlands Statistician, SAS consultant home.hccnet.nl/jim.groeneveld

On Thu, 17 Aug 2006 05:55:24 -0400, Minze Su <slhappyls@GMAIL.COM> wrote:

>Hi all, > > I have a data set having the following variables(id,v1,v2). After some >exploratory data analysis, I kick out 9 outliers. The old data set is >called ¡®OLD¡¯£¬and cleaned one is called 'NEW'. Now I want to know the id >of those 9 outliers. How can I do this? Thank you in advance! > >BR, > >Minze