On Fri, 26 May 2006 16:28:20 +0000, toby dunn <tobydunn@HOTMAIL.COM> wrote:

>Talbot , > >Uhhh yeah you are, you are letting SQL do an implicit conversion from >numeric to character for you when it shoves the summed value into the macro >var. It uses best12 formatting and normally this isnt a bad thing and works

I think BEST12 would have worked (barely) for TMK in this case, but as I recall SQL for some reason actually uses BEST8.

>fine. However when you have a large number it does truncate things or when >you have decimals and the number as a whole is large enough things get >dropped off. SO use a put function and put the variable into the macro var >with the right numeric format. > > >Data one ; > x = 16610879.18 ; > put x= ; >run ; > >Proc sql noprint ; >select x into : XVar >from one ; >quit ; > >%put %sysfunc( putn( &XVar , 12.2 ) ) ; > > >Proc sql noprint ; >select put( x , 12.2 ) into : XVar >from one ; >quit ; > >%put %sysfunc( putn( &XVar , 12.2 ) ) ; > > > >Toby Dunn > > > > > >From: Talbot Michael Katz <topkatz@MSN.COM> >Reply-To: Talbot Michael Katz <topkatz@MSN.COM> >To: SAS-L@LISTSERV.UGA.EDU >Subject: PROC SQL Sum Into Macro Variable >Date: Fri, 26 May 2006 12:12:21 -0400 > >Hi. > >Please have patience with me. I think I asked a question like this >before, but somehow I still haven't gotten it. > >When I sum into macro variable as follows: > >proc sql ; > reset noprint ; > select sum(x * (x ge &lim.)) > into :sum_x_above_lim > from ds1 > ; >quit ; > >310 %put sum_x_above_lim = >310! %sysfunc(putn(&sum_x_above_lim.,comma15.2)) ; >sum_x_above_lim = 16,610,879.00 > >316 %put sum_x_above_lim = &sum_x_above_lim. ; >sum_x_above_lim = 16610879 > > >When I sum the same into a data set: > >proc sql ; > create table sum_x_above_lim as > select sum(x * (x ge &lim.)) as > sum_x_above_lim > from ds1 > ; >quit ; > >and I print out the value I get: 16610879.18 > > >Why am I losing the part after the decimal point in the macro variable? > > >-- TMK -- >"The Macro Klutz"