| Date: | Wed, 27 Jan 1999 16:55:18 -0800 |
|---|
| Reply-To: | Tom Metzler <metzler@ITSA.UCSF.EDU> |
|---|
| Sender: | "SAS(r) Discussion" <SAS-L@UGA.CC.UGA.EDU> |
|---|
| From: | Tom Metzler <metzler@ITSA.UCSF.EDU> |
|---|
| Subject: | Merging datasets with missing values |
|---|
| Content-Type: | multipart/mixed;
|
|---|
I want to do a match-merge of two SAS datasets that have identical variable
names.
The problem is that a given observation may have a missing value for a
variable in
one set, and a non-missing value in the other set. I want to include the
non-missing
value in the output set, no matter which input file it comes from. (If both
sets
contain non-missing values, they will be identical, so it doesn't matter
which set
they are taken from). Is there a straightforward way of doing this? The
following
illustrates what I'm trying to do:
dataset1:
id var1 var2 var3
1 1 . 3
2 4 5 .
3 7 8 9
dataset2:
id var1 var2 var3
1 1 2 .
3 7 . 9
4 . 11 12
merged dataset:
id var1 var2 var3
1 1 2 3
2 4 5 .
3 7 8 9
4 . 11 12
-----------------------------
Tom Metzler
PTSD Research Program (116P)
San Francisco VA Medical Center
4150 Clement St.
San Francisco, CA 94121
(415) 221-4810 x3082
winmail.dat [application/ms-tnef]
| 
