Date:         Thu, 30 Jul 2009 07:23:37 -0500
Reply-To:     "Data _null_;" <iebupdte@GMAIL.COM>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         "Data _null_;" <iebupdte@GMAIL.COM>
Subject:      Re: Normalising data
Comments: To: "Chipulu M." <M.CHIPULU@soton.ac.uk>
In-Reply-To:  <ce1fb7450907300513j3eb878cfj770df9c38dd6965e@mail.gmail.com>
Content-Type: text/plain; charset=ISO-8859-1

I'm think you want to "standardize" the values of PROB1-PROB4 (by
GROUP).  If that is the case I think the following PROC STDIZE
METHOD=SUM is what you are looking for.

data max;
   infile cards dsd;
   input Group Pred Prob1-Prob4;
   cards;
1,1,0.238309,0.19227,0.146976,0.111108
1,2,0.05607,0.069471,0.080536,0.089676
1,3,0.09679,0.108928,0.113211,0.112122
1,4,0.024849,0.033166,0.042181,0.052461
1,5,0.02336,0.031289,0.039979,0.050014
2,5,0.386618,0.217088,0.129933,0.083083
2,5,0.035242,0.045883,0.056527,0.067598
2,5,0.072332,0.086235,0.095565,0.101228
2,5,0.013512,0.018532,0.02443,0.031815
3,5,0.040344,0.051891,0.062961,0.073925
3,5,0.091758,0.104496,0.110011,0.110393
3,5,0.021055,0.028359,0.036498,0.046079
3,1,0.082671,0.096185,0.103687,0.106639
4,2,0.066602,0.080488,0.090602,0.097623
4,3,0.113718,0.122987,0.122551,0.116398
4,4,0.060518,0.074196,0.084942,0.093256
4,5,0.176115,0.164532,0.142315,0.118834
5,5,0.020157,0.027208,0.035115,0.044493
5,5,0.017026,0.023154,0.030182,0.038734
5,5,0.092988,0.10559,0.110813,0.110837
5,5,0.057829,0.071352,0.082306,0.091133
6,5,0.152136,0.150355,0.137,0.119706
6,5,0.038345,0.049555,0.060486,0.071525
6,5,0.018502,0.025073,0.03253,0.041496
;;;;
   run;
proc stdize method=sum;
   by group;
   var prob:;
   run;
proc print;
   run;
proc means n sum;
   class group;
   var prob:;
   run;

On 7/30/09, Data _null_; <iebupdte@gmail.com> wrote:
> The data you posted difficult to read.  Does it look like that posted
> below.  It is best to include a bit of code to read any data you post
> to SAS-L.
>
> data max;
>   infile cards dsd;
>   input Group Pred Prob1-Prob4;
>   cards;
> 1,1,0.238309,0.19227,0.146976,0.111108
> 1,2,0.05607,0.069471,0.080536,0.089676
> 1,3,0.09679,0.108928,0.113211,0.112122
> 1,4,0.024849,0.033166,0.042181,0.052461
> 1,5,0.02336,0.031289,0.039979,0.050014
> 2,5,0.386618,0.217088,0.129933,0.083083
> 2,5,0.035242,0.045883,0.056527,0.067598
> 2,5,0.072332,0.086235,0.095565,0.101228
> 2,5,0.013512,0.018532,0.02443,0.031815
> 3,5,0.040344,0.051891,0.062961,0.073925
> 3,5,0.091758,0.104496,0.110011,0.110393
> 3,5,0.021055,0.028359,0.036498,0.046079
> 3,1,0.082671,0.096185,0.103687,0.106639
> 4,2,0.066602,0.080488,0.090602,0.097623
> 4,3,0.113718,0.122987,0.122551,0.116398
> 4,4,0.060518,0.074196,0.084942,0.093256
> 4,5,0.176115,0.164532,0.142315,0.118834
> 5,5,0.020157,0.027208,0.035115,0.044493
> 5,5,0.017026,0.023154,0.030182,0.038734
> 5,5,0.092988,0.10559,0.110813,0.110837
> 5,5,0.057829,0.071352,0.082306,0.091133
> 6,5,0.152136,0.150355,0.137,0.119706
> 6,5,0.038345,0.049555,0.060486,0.071525
> 6,5,0.018502,0.025073,0.03253,0.041496
> ;;;;
>   run;
> proc print;
>   run;
>
> On 7/30/09, Chipulu M. <M.CHIPULU@soton.ac.uk> wrote:
> > I have some data a sample of which looks like this:
>