Date:         Wed, 5 Jan 2011 22:26:08 -0800
Reply-To:     Jack Hamilton <jfh@STANFORDALUMNI.ORG>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         Jack Hamilton <jfh@STANFORDALUMNI.ORG>
Subject:      Re: PROC GCHART: Explicitly control the ordering of group
              variables
Comments: To: Matthew Pettis <matt.pettis@THOMSONREUTERS.COM>
In-Reply-To:  <201101060603.p05LJElR004955@wasabi.cc.uga.edu>
Content-Type: text/plain; charset=us-ascii

I haven't used GCHART in a long time, but will GAXIS with ORDER= do what you want?


On Jan 5, 2011, at 10:03 PM, Matthew Pettis wrote:

> Hi,
>
> I am trying to make a gchart in which I explictly control the order of the
> group= option in the code.
>
> I have a solution, and I am wondering if there is a better way than how I
> did it.
>
> My solution was to explicitly add a numeric order varialbe to my dataset
> that represents the order in which I want the grouping values to display,
> and then apply a format with the values that I want displayed back onto the
> newly created order variable.  This is in the code below.
>
> This to me seems sub-optimal, and seems like a request that would be so
> common as to have some option in gchart that would aid me.  I thought of
> using 'midpoints', but I included that example in the code below and, if you
> run it, you can see it does not do what I want.
>
> Any "yer doin' it wrong," reply, with a better solution, is heartily
> welcomed.
>
> TIA,
> Matt
>
>
>
>
>
>    *
>    --------------------------------------------------------------------
>    Code Section
>    --------------------------------------------------------------------
>    ;
>
> data have;
>    infile datalines;
>    input X:$1. Y$1. Z order_val;
> datalines;
> A E 90 1
> A D 70 2
> A F 50 3
> A G 30 4
> A H 10 5
> B E 95 1
> B D 65 2
> B F 55 3
> B G 25 4
> B H 15 5
> ;
> run;
>
>    * This is close to what I want;
>    * It has the right grouping and subgrouping.;
>    * But I want to be able to control the order;
>    * In a general descending manner:;
> goptions reset=all;
> proc gchart data=have;
>    vbar X /
>        sumvar=Z
>        group=Y
>        subgroup=X
>    ;
>    run;
> quit;
>
>
>
>
>
>    * I do have a solution to the problem,;
>    * which is to do the following:;
> proc format;
>    value morder
>        1 = 'E'
>        2 = 'D'
>        3 = 'F'
>        4 = 'G'
>        5 = 'H'
>    ;
> run;
>
> proc datasets;
>    modify have;
>    format order_val morder.;
>    run;
> quit;
>
> goptions reset=all;
> proc gchart data=have;
>    vbar X /
>        sumvar=Z
>        group=order_val
>        subgroup=X
>    ;
>    run;
> quit;
>
>
>
>
>
>    * I thought I could control the;
>    * order with something simple;
>    * like the following:;
>    *   midpoints='E' 'D' 'F' 'G' 'H';
>    * but you can run the code below;
>    * and see that it does not work out.;
>    * BAD/UNWANTED OUTPUT:;
> goptions reset=all;
> proc gchart data=have;
>    vbar X /
>        sumvar=Z
>        group=Y
>        subgroup=X
>        midpoints='E' 'D' 'F' 'G' 'H'
>    ;
>    run;
> quit;