# st: How to sum group of variables to a single variable in a large data set?

 From To Subject st: How to sum group of variables to a single variable in a large data set? Date Mon, 5 Mar 2007 13:35:18 +0200

```Hi

I wanted to ask how do I sum group of variables into a single variable
in a large data set?

Suppose I have list of variables as follows:

diag3201

diag4201

diag8203

diag9201

diag9202

diag9203

diag9204

diag10214

diag10218

diag11201

diag98201

diag98202

diag98204

diag98205

diag98206

diag98207

diag100201

diag100202

diag100203

diag100204

diag100205

diag100206

diag100207

diag100208

diag100209

diag100210

diag100211

diag100212

diag100213

diag100214

diag100215

diag100216

diag100217

diag100218

diag100219

diag100220

diag100221

The last three digits for each variable are the same; starts with 201
and end with 222 for some of them. The first digits indicate a
diagnosis. For instance, for diagnosis 98 we have variables diag98201
till diag98207, for diagnosis 9 we have variables diag9201-diag9204. I
want to generate new variables for the diagnoses that equal to the sum
of the variables that relate to this diagnosis. For example I want to
generate diag98=diag98201+ diag98202+ …+ diag98207 or diag9=diag9201+
diag9202+..diag9207.

I know how to do it for one diagnosis at a time, I use egen diag98=
rsum(diag98201-diag98207), but I need to do it for a large number of
diagnoses so how can I do it for all of them?

Thanks

David

p.s. I have stata8

﻿﻿﻿﻿﻿﻿﻿﻿﻿﻿﻿
David Messika
The Health Economics Unit
The Gertner Institute for Epidemiology and Health Policy Research
Chaim Sheba Medical Center
Ramat-Gan, Israel
(: 972-3-5303935 (: 972-3-5303277

*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/
```