On 17 Feb, 08:51, Metricsdude <bsj_ad...@hotmail.com> wrote: > On 16 Feb, 16:57, chang_y_ch...@HOTMAIL.COM (Chang Chung) wrote: > > > > > > > On Mon, 16 Feb 2009 08:33:01 -0800, Metricsdude <bsj_ad...@HOTMAIL.COM> > > wrote: > > ... > > > >Any advice would be much appreaciated as i would like to learn the > > >best way to do this as i want to add in more complexity to this.... > > > hi, > > so, the goal is to create a data set, mergelookup, which has last month's > > dates as observations, each date repeated six times, each time with a > > different bill code -- GR1, GR2,..., GR6. If that is the goal, then I think > > there are much easier ways. here is one. hth. > > cheers, > > chang > > > %*-- first and last day of last month --*; > > %let bdt = %sysfunc(intnx(month,%sysfunc(date()),-1,b),date9.); > > %let edt = %sysfunc(intnx(month,%sysfunc(date()),-1,e),date9.); > > > /* create the mergelookup table */ > > data lookup; > > attrib billcode length=$3 day format=date9.; > > do billcode = "GR1", "GR2", "GR3", "GR4", "GR5", "GR6"; > > do day = "&bdt"d to "&edt"d; > > output; > > end; > > end; > > run; > > Sorry i have another probably stupid question... > > I now want to repeat all these observations for various market > participants held in another dataset (called MAP) which will vary each > month would this now be the appropriate time to use a macro or do you > have another simple solution that could be applied????- Hide quoted text - > > - Show quoted text -

Simply modify the previous example, get each unique participant, create a macro variable and iterate over it (assumes a length of 6 for participant in MAP dataset */

proc freq data=map noprint ; table participant / out=f_part ; run ;

proc sql ; select quote(particpant) into :PART_LIST separated by ',' from f_part ; quit ;

data lookup; attrib part $6. billcode length=$3 day format=date9.; do part = &PART_LIST ; do billcode = "GR1", "GR2", "GR3", "GR4", "GR5", "GR6"; do day = "&bdt"d to "&edt"d; output; end; end; end ; run;