rushi.b.patel@GMAIL.COM wrote back: >Thanks for your valuable comments and the code. It works fine. I can >automate a lot of stuff with that and is more flexible than the clumsy >code that I coded. Plus a valuable lesson in Macro Design is learned. > >One question though.. > >Is there a one pass solution if I am interested in having percentage >of weighted missings.. > >For example, I have added a wt variable to the dataset that you use to >illustrate. Instead of outputing the percent of missing number of >observations, I am interested in percent missing by wt variable. So >the output for variable x should be 0.50 and not 0.25.. > >All I can think of is a clumsy way requiring multiple passes depending >upon the number of variables.. > >data w2003 ( drop = x ) w2004 ( drop = c ) ; > input x y z a $ b $ c $ wt; > cards ; > 1 2 3 a b c 5 > . . . . . . 10 > 2 . 4 . a . 4 > 7 . . . c c 1 > run ;

What do the weights represent? Why are you introducing them?

David -- David L. Cassell mathematical statistician Design Pathways 3115 NW Norwood Pl. Corvallis OR 97330

_________________________________________________________________ Don’t just search. Find. Check out the new MSN Search! http://search.msn.click-url.com/go/onm00200636ave/direct/01/