LISTSERV at the University of Georgia
Menubar Imagemap
Home Browse Manage Request Manuals Register
Previous messageNext messagePrevious in topicNext in topicPrevious by same authorNext by same authorPrevious page (May 2007, week 4)Back to main SAS-L pageJoin or leave SAS-L (or change settings)ReplyPost a new messageSearchProportional fontNon-proportional font
Date:         Fri, 25 May 2007 09:57:51 -0500
Reply-To:     Tom White <tw2@MAIL.COM>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         Tom White <tw2@MAIL.COM>
Subject:      SAS Data Step--Fill in missing values
Content-Type: text/plain; charset="iso-8859-1"

data foo; input A $ 1-10 B $ 11-20 C $ 21-30 D best12.; cards; 12 5B231 0 56 1003 0 19 7AS125 0 19 7AS125 0 19 7AS125 1260 1 19 7AS125 1260 1 32 909 0 65 3012 0 60 1601 0 60 1601 0 1601 6597 1 60 1601 6597 1 95 7AH014 0 95 7AH014 0 95 7AH014 838 1 95 7AH014 838 2 95 7AH014 838 3 95 7AH014 838 4 95 7AH014 838 1 95 7AH014 838 2 95 7AH014 838 3 95 7AH014 838 4 71 1518 0 71 1518 0 1518 6956 1 1518 6956 2 1518 6956 3 71 1518 6956 1 71 1518 6956 2 71 1518 6956 3 61 AS009 0 61 AS009 0 61 AS009 0 61 AS009 0 33 18541 0 33 18541 0 33 18541 0 66 B2741 0 86 2246 0 86 2246 0 2246 07436 1 2246 07436 2 2246 07436 3 2246 07436 4 86 2246 07436 1 86 2246 07436 2 86 2246 07436 3 13 264 0 87 2304 0 14 7275 0 25 110 0 25 110 0 110 1774 1 25 110 1774 1 run;

Hello SAS-L,

I have selected a few cases from a larger dataset to present my problem.

This data is simply member/customer account level data in health insurance industry.

I would like to fill in the missing values for variables A and C.

(Variables B and D is never missing.)

If you look at the data observe that:

A and C have some missing values.

EXAMPLES:

1. A=12 has a missing C. There is just no way to fill in this case because we don't have enough information. Therefore we leave this one as is.

2. A=56 Same as 1.

3. A=19. Here, we can easily infer that C should be 1260. (Looking also at D column might be of help.) Therefore, the first two observations with A=56 should take on the value 1260 for C.

And so on.

4. Let's look at A=60. Here we have both A and C missing. Again, you can easily see that the third missing obs of A=60 should be 60 (again notice how D is not 0, but a value other than 0, in this case, 1. Just mentioning this if it is of any help.) and that the missing C should be 6597.

5.Similarly: A=71 should be filled in with 71 and C should be filled in with 6956

A=86 should be filled in with 86 and C should be filled in with 07436

A=25 should be filled in with 25 and C should be filled in with 1774

I have SAS v8.2

Thank you.

tom

= Affordable Radiant Heat Do It Yourself Radiant heating packages, diy install. Expert Design assistance free. http://a8-asy.a8ww.net/a8-ads/adftrclick?redirectid=bdfab744df8e9c8bd3dbadb220614ffd


Back to: Top of message | Previous page | Main SAS-L page