topic Re: how to program like this? in SAS Procedures

topic Re: how to program like this? in SAS Procedures https://communities.sas.com/t5/SAS-Procedures/how-to-program-like-this/m-p/107342#M29863 <HTML><HEAD></HEAD><BODY><P>Hi Mike,</P><P>If you are going to try Astounding's code you have to try my code(I learned from Ksharp) as well:smileysilly:.</P></BODY></HTML> Thu, 30 Aug 2012 01:31:28 GMT Linlin 2012-08-30T01:31:28Z how to program like this? https://communities.sas.com/t5/SAS-Procedures/how-to-program-like-this/m-p/107332#M29853 <HTML><HEAD></HEAD><BODY><P>Is there one function which can return a value stored in the next observation on the opposite of  LAG function ?For example,</P><P>data have;</P><P>input x;</P><P>datalines;</P><P>1</P><P>2</P><P>3</P><P>4</P><P>5</P><P>6</P><P>7</P><P>8</P><P>9</P><P>10</P><P>;</P><P>run;</P><P></P><P>I want to assign the average value of the next 5 observations to one new variable. How to achieve this? Thanks in advance.</P></BODY></HTML> Wed, 29 Aug 2012 11:07:47 GMT https://communities.sas.com/t5/SAS-Procedures/how-to-program-like-this/m-p/107332#M29853 TomiKong 2012-08-29T11:07:47Z Re: how to program like this? https://communities.sas.com/t5/SAS-Procedures/how-to-program-like-this/m-p/107333#M29854 <HTML><HEAD></HEAD><BODY><P>How big is your dataset?  If it is not horribly long, try PROC SORT with a descending option and an out=newdata, and then apply the mean function to the lagged variables in newdata.  I know there was something about "look ahead" functions in a thread recently, but I have made this method work before.</P><P></P><P>Steve Denham</P></BODY></HTML> Wed, 29 Aug 2012 12:00:01 GMT https://communities.sas.com/t5/SAS-Procedures/how-to-program-like-this/m-p/107333#M29854 SteveDenham 2012-08-29T12:00:01Z Re: how to program like this? https://communities.sas.com/t5/SAS-Procedures/how-to-program-like-this/m-p/107334#M29855 <HTML><HEAD></HEAD><BODY><P>What do you want to assign to records 6 thru 10?</P><P></P><P>The look forward method that Steve mentioned may have been the one that uses multiple set statements.  e.g.:</P><P></P><P>data want;</P><P>  set have;</P><P>  set have ( firstobs = 2 keep = x rename = (x = _x1) )</P><P>      have (      obs = 1 drop = _all_               );</P><P>  set have ( firstobs = 3 keep = x rename = (x = _x2) )</P><P>      have (      obs = 1 drop = _all_               )</P><P>      have (      obs = 1 drop = _all_               );</P><P>  set have ( firstobs = 4 keep = x rename = (x = _x3) )</P><P>      have (      obs = 1 drop = _all_               )</P><P>      have (      obs = 1 drop = _all_               )</P><P>      have (      obs = 1 drop = _all_               );</P><P>  set have ( firstobs = 5 keep = x rename = (x = _x4) )</P><P>      have (      obs = 1 drop = _all_               )</P><P>      have (      obs = 1 drop = _all_               )</P><P>      have (      obs = 1 drop = _all_               )</P><P>      have (      obs = 1 drop = _all_               );</P><P>  set have ( firstobs = 6 keep = x rename = (x = _x5) )</P><P>      have (      obs = 1 drop = _all_               )</P><P>      have (      obs = 1 drop = _all_               )</P><P>      have (      obs = 1 drop = _all_               )</P><P>      have (      obs = 1 drop = _all_               )</P><P>      have (      obs = 1 drop = _all_               );</P><P>  mean_x=mean(of _x:);</P><P>run;</P><P></P><P>I think that would be the most efficient method.  Conversely, if you have proc expand, you could use it to expand the dataset, and then use another datastep to calculate the mean.  e.g.:</P><P></P><P>proc expand data=have out=want2 (drop=time);  </P><P>   convert x=x1  / method=none transformout=(lead 1);  </P><P>   convert x=x2  / method=none transformout=(lead 2);  </P><P>   convert x=x3  / method=none transformout=(lead 3);  </P><P>   convert x=x4  / method=none transformout=(lead 4);  </P><P>   convert x=x5  / method=none transformout=(lead 5);  </P><P>run;</P></BODY></HTML> Wed, 29 Aug 2012 12:29:22 GMT https://communities.sas.com/t5/SAS-Procedures/how-to-program-like-this/m-p/107334#M29855 art297 2012-08-29T12:29:22Z Re: how to program like this? https://communities.sas.com/t5/SAS-Procedures/how-to-program-like-this/m-p/107335#M29856 <HTML><HEAD></HEAD><BODY><P>OK, here's my shot at it.  I assumed that when we get to observation 6, and there are only 4 foward-looking observations, that we should take the mean of those 4:</P><P></P><P><SPAN style="font-size: 8pt;"><SPAN lang="EN">  </SPAN></SPAN></P><P>data test;</P><P><SPAN style="font-size: 8pt;"><SPAN lang="EN"></SPAN></SPAN></P><P>do i=1 to 10;</P><P><SPAN style="font-size: 8pt;"><SPAN lang="EN"></SPAN></SPAN></P><P>output;</P><P><SPAN style="font-size: 8pt;"><SPAN lang="EN"></SPAN></SPAN></P><P>end;</P><P><SPAN style="font-size: 8pt;"><SPAN lang="EN"></SPAN></SPAN></P><P>run;</P><P><SPAN style="font-size: 8pt;"><SPAN lang="EN"></SPAN></SPAN></P><P>data test2;</P><P><SPAN style="font-size: 8pt;"><SPAN lang="EN"></SPAN></SPAN></P><P>set test nobs=_nobs_;</P><P><SPAN style="font-size: 8pt;"><SPAN lang="EN"></SPAN></SPAN></P><P>array forward {0:4} forward1-forward5;</P><P><SPAN style="font-size: 8pt;"><SPAN lang="EN"></SPAN></SPAN></P><P>retain forward1-forward5;</P><P><SPAN style="font-size: 8pt;"><SPAN lang="EN"></SPAN></SPAN></P><P>if _n_=1 then do _j_=2 to 5;</P><P><SPAN style="font-size: 8pt;"><SPAN lang="EN"></SPAN></SPAN></P><P>set test (rename=(i=forward_value)) point=_j_;</P><P><SPAN style="font-size: 8pt;"><SPAN lang="EN"></SPAN></SPAN></P><P>forward{mod(_j_,5)}=forward_value;</P><P><SPAN style="font-size: 8pt;"><SPAN lang="EN"></SPAN></SPAN></P><P>end;</P><P><SPAN style="font-size: 8pt;"><SPAN lang="EN"></SPAN></SPAN></P><P>if _n_+5 <= _nobs_ then do;</P><P><SPAN style="font-size: 8pt;"><SPAN lang="EN"></SPAN></SPAN></P><P>_j_=_n_+5;</P><P><SPAN style="font-size: 8pt;"><SPAN lang="EN"></SPAN></SPAN></P><P>set test (rename=(i=forward_value)) point=_j_;</P><P><SPAN style="font-size: 8pt;"><SPAN lang="EN"></SPAN></SPAN></P><P>forward{mod(_n_,5)}=forward_value;</P><P><SPAN style="font-size: 8pt;"><SPAN lang="EN"></SPAN></SPAN></P><P>end;</P><P><SPAN style="font-size: 8pt;"><SPAN lang="EN"></SPAN></SPAN></P><P>else forward{mod(_n_,5)}=.;</P><P><SPAN style="font-size: 8pt;"><SPAN lang="EN"></SPAN></SPAN></P><P>forward_5_mean = mean(of forward{*});</P><P><SPAN style="font-size: 8pt;"><SPAN lang="EN"></SPAN></SPAN></P><P>put _all_;</P><P><SPAN style="font-size: 8pt;"><SPAN lang="EN"></SPAN></SPAN></P><P>run;</P><P><SPAN style="font-size: 8pt;"><SPAN lang="EN"></SPAN></SPAN></P></BODY></HTML> Wed, 29 Aug 2012 13:24:00 GMT https://communities.sas.com/t5/SAS-Procedures/how-to-program-like-this/m-p/107335#M29856 Astounding 2012-08-29T13:24:00Z Re: how to program like this? https://communities.sas.com/t5/SAS-Procedures/how-to-program-like-this/m-p/107336#M29857 <HTML><HEAD></HEAD><BODY><P>data have;<BR />do var=1 to 10;<BR />output;<BR />end;<BR />run;<BR />data temp;<BR />array v(0:4) _temporary_;<BR />set have;<BR />v(mod(_n_,5))=var;<BR />if _n_>=6 then do; mean_5=mean( of v(*)); output;end;<BR />keep mean_5;<BR />run;<BR />data want;<BR />  merge have temp;<BR /> proc print;run;<BR />                                       obs    var    mean_5</P><P>                                         1      1       4<BR />                                         2      2       5<BR />                                         3      3       6<BR />                                         4      4       7<BR />                                         5      5       8<BR />                                         6      6       .<BR />                                         7      7       .<BR />                                         8      8       .<BR />                                         9      9       .<BR />                                        10     10       .</P></BODY></HTML> Wed, 29 Aug 2012 14:28:23 GMT https://communities.sas.com/t5/SAS-Procedures/how-to-program-like-this/m-p/107336#M29857 Linlin 2012-08-29T14:28:23Z Re: how to program like this? https://communities.sas.com/t5/SAS-Procedures/how-to-program-like-this/m-p/107337#M29858 <HTML><HEAD></HEAD><BODY><P>Shamelessly stealing Linlin' idea, but modified to finish in one data step:</P><P></P><P>data have;</P><P>do x=1 to 10;</P><P>output;   </P><P>end;</P><P>run;</P><P></P><P>data want (drop=x rename=y=x) ;</P><P>array nx5(0:5) _temporary_;</P><P>set have end=last nobs=nobs;</P><P>nx5(mod(_n_,6))=x;</P><P>if _n_>5 then do;</P><P>   y=nx5(mod(_n_-5,6));</P><P>   avg=(sum(of nx5(*))-y)/(n(of nx5(*))-1);</P><P>   output;</P><P>end;</P><P>if last then do ;</P><P>       do _n_=nobs+1 to nobs+5;</P><P>          y=nx5(mod(_n_-5,6));</P><P>       call missing(nx5(mod(_n_-5,6)),nx5(mod(_n_-6,6)));</P><P>       avg=mean(of nx5(*));</P><P>       output;</P><P>       end;</P><P></P><P>end;</P><P>run;</P><P></P><P>proc print;run;</P><P></P><P>Haikuo</P></BODY></HTML> Wed, 29 Aug 2012 17:03:29 GMT https://communities.sas.com/t5/SAS-Procedures/how-to-program-like-this/m-p/107337#M29858 Haikuo 2012-08-29T17:03:29Z Re: how to program like this? https://communities.sas.com/t5/SAS-Procedures/how-to-program-like-this/m-p/107338#M29859 <HTML><HEAD></HEAD><BODY><P>hi ... another idea</P><P></P><P>though it's grossly inefficient to keep reading five new values from the data set over-and-over again, </P><P>it took < 1.5 seconds to process a million observations (on my "not so fast" PC)</P><P></P><P><SPAN style="font-family: 'courier new', courier;"><STRONG>data want (keep=x mean);</STRONG></SPAN></P><P><SPAN style="font-family: 'courier new', courier;"><STRONG>set have nobs=obs;</STRONG></SPAN></P><P><SPAN style="font-family: 'courier new', courier;"><STRONG>if (obs - _n_) ge 5 then do;</STRONG></SPAN></P><P><SPAN style="font-family: 'courier new', courier;"><STRONG>   do nrec=_n_ + 1 to _n_ + 5;</STRONG></SPAN></P><P><SPAN style="font-family: 'courier new', courier;"><STRONG>      set have (rename=(x=y)) point=nrec;</STRONG></SPAN></P><P><SPAN style="font-family: 'courier new', courier;"><STRONG>      five = sum(five, y);</STRONG></SPAN></P><P><SPAN style="font-family: 'courier new', courier;"><STRONG>   end;</STRONG></SPAN></P><P><SPAN style="font-family: 'courier new', courier;"><STRONG>   mean = five / 5;</STRONG></SPAN></P><P><SPAN style="font-family: 'courier new', courier;"><STRONG>end;</STRONG></SPAN></P><P><SPAN style="font-family: 'courier new', courier;"><STRONG>else call missing (mean);</STRONG></SPAN></P><P><SPAN style="font-family: 'courier new', courier;"><STRONG>run;</STRONG></SPAN></P></BODY></HTML> Wed, 29 Aug 2012 22:36:15 GMT https://communities.sas.com/t5/SAS-Procedures/how-to-program-like-this/m-p/107338#M29859 MikeZdeb 2012-08-29T22:36:15Z Re: how to program like this? https://communities.sas.com/t5/SAS-Procedures/how-to-program-like-this/m-p/107339#M29860 <HTML><HEAD></HEAD><BODY><P>Mike,</P><P></P><P>Since this is a more automated version of the code that I originally proposed (okay .. I stole it from Howard's sasCommunity.or Look Ahead Look Back article), I think that you really want to include a keep option in the additional set statements.</P></BODY></HTML> Wed, 29 Aug 2012 22:59:12 GMT https://communities.sas.com/t5/SAS-Procedures/how-to-program-like-this/m-p/107339#M29860 art297 2012-08-29T22:59:12Z Re: how to program like this? https://communities.sas.com/t5/SAS-Procedures/how-to-program-like-this/m-p/107340#M29861 <HTML><HEAD></HEAD><BODY><P>hi Art ... I was assuming the one variable data set that was in the original question </P></BODY></HTML> Wed, 29 Aug 2012 23:05:12 GMT https://communities.sas.com/t5/SAS-Procedures/how-to-program-like-this/m-p/107340#M29861 MikeZdeb 2012-08-29T23:05:12Z Re: how to program like this? https://communities.sas.com/t5/SAS-Procedures/how-to-program-like-this/m-p/107341#M29862 <HTML><HEAD></HEAD><BODY><P>Mike,</P><P></P><P>Did you try my version as well?  It looks longer, but it reads even less data.  It recognizes that 4 of the 5 records that you are reading with point= don't have to be read.  They were just read on the previous observation, and can be retained instead.</P><P></P><P>Just remember to add keep= , and remove the put statement at the end.</P><P></P><P>Astounding</P></BODY></HTML> Thu, 30 Aug 2012 00:20:39 GMT https://communities.sas.com/t5/SAS-Procedures/how-to-program-like-this/m-p/107341#M29862 Astounding 2012-08-30T00:20:39Z Re: how to program like this? https://communities.sas.com/t5/SAS-Procedures/how-to-program-like-this/m-p/107342#M29863 <HTML><HEAD></HEAD><BODY><P>Hi Mike,</P><P>If you are going to try Astounding's code you have to try my code(I learned from Ksharp) as well:smileysilly:.</P></BODY></HTML> Thu, 30 Aug 2012 01:31:28 GMT https://communities.sas.com/t5/SAS-Procedures/how-to-program-like-this/m-p/107342#M29863 Linlin 2012-08-30T01:31:28Z Re: how to program like this? https://communities.sas.com/t5/SAS-Procedures/how-to-program-like-this/m-p/107343#M29864 <HTML><HEAD></HEAD><BODY><P>For your question, I would like to use two skill: self-merge or SQL .</P><P>I prefer to SQL , because it might be faster when your table is large. Just a guess .</P><P></P><P></P><P></P><PRE>data have; input x; datalines; 1 2 3 4 5 6 7 8 9 10 ; run; /*self-merge skill*/ data want; merge have have(firstobs=2 rename=(x=x1)) have(firstobs=3 rename=(x=x2)) have(firstobs=4 rename=(x=x3)) have(firstobs=5 rename=(x=x4))  have(firstobs=6 rename=(x=x5)); avg=mean(of x1-x5); drop x1-x5; run; /*SQL skill*/ data have; set have; n=_n_; run; proc sql; create table want(drop=n) as   select *,(select mean(x) from have where n between a.n+1 and a.n+5) as avg    from have as a; quit; </PRE><P></P><P></P><P></P><P></P><P></P><P>Ksharp</P></BODY></HTML> Thu, 30 Aug 2012 02:24:06 GMT https://communities.sas.com/t5/SAS-Procedures/how-to-program-like-this/m-p/107343#M29864 Ksharp 2012-08-30T02:24:06Z Re: how to program like this? https://communities.sas.com/t5/SAS-Procedures/how-to-program-like-this/m-p/107344#M29865 <HTML><HEAD></HEAD><BODY><P>OK, I found some time.  I tested my code vs. Mike's and found that mine took less than 50% of the time.</P><P></P><P>With apologies, I didn't test other approaches but I know from past experience that MERGE takes much longer than SET.</P></BODY></HTML> Thu, 30 Aug 2012 18:11:56 GMT https://communities.sas.com/t5/SAS-Procedures/how-to-program-like-this/m-p/107344#M29865 Astounding 2012-08-30T18:11:56Z Re: how to program like this? https://communities.sas.com/t5/SAS-Procedures/how-to-program-like-this/m-p/107345#M29866 <HTML><HEAD></HEAD><BODY><P>hi ... I did say "<SPAN style="font-family: 'Helvetica Neue', Helvetica, Arial, 'Lucida Grande', sans-serif; font-size: 12.800000190734863px; background-color: #ffffff;">though it's grossly inefficient</SPAN> ..."</P><P></P><P>so I modified my posting to do the same thing as Astounding's (his better idea of just remove one value and add one value to the total of X across five observations) ...</P><P></P><P><SPAN style="font-family: 'courier new', courier;"><STRONG>data want (keep=x mean);</STRONG></SPAN></P><P><SPAN style="font-family: 'courier new', courier;"><STRONG>* add first five values;</STRONG></SPAN></P><P><SPAN style="font-family: 'courier new', courier;"><STRONG>do nrec=1 to 5;</STRONG></SPAN></P><P><SPAN style="font-family: 'courier new', courier;"><STRONG>   set have nobs=obs;</STRONG></SPAN></P><P><SPAN style="font-family: 'courier new', courier;"><STRONG>   five = sum(five,x);</STRONG></SPAN></P><P><SPAN style="font-family: 'courier new', courier;"><STRONG>end;</STRONG></SPAN></P><P><SPAN style="font-family: 'courier new', courier;"><STRONG>* drop one value, add one value;</STRONG></SPAN></P><P><SPAN style="font-family: 'courier new', courier;"><STRONG>do nrec=6 by 1 until(done);</STRONG></SPAN></P><P><SPAN style="font-family: 'courier new', courier;"><STRONG>   set have end=done;</STRONG></SPAN></P><P><SPAN style="font-family: 'courier new', courier;"><STRONG>   set have (keep=x rename=(x=y)) point=nrec;</STRONG></SPAN></P><P><SPAN style="font-family: 'courier new', courier;"><STRONG>   five = five - x + y;</STRONG></SPAN></P><P><SPAN style="font-family: 'courier new', courier;"><STRONG>   mean = five / 5;</STRONG></SPAN></P><P><SPAN style="font-family: 'courier new', courier;"><STRONG>   if nrec gt obs then call missing(mean);</STRONG></SPAN></P><P><SPAN style="font-family: 'courier new', courier;"><STRONG>   output;</STRONG></SPAN></P><P><SPAN style="font-family: 'courier new', courier;"><STRONG>end;</STRONG></SPAN></P><P><SPAN style="font-family: 'courier new', courier;"><STRONG>stop;</STRONG></SPAN></P><P><SPAN style="font-family: 'courier new', courier;"><STRONG>run;</STRONG></SPAN></P><P></P><P>I tried the above versus Astounding's code with various size data sets (up to 1 million obs) and they take the same amount of time (with 1 million observations, each takes about 0.5 seconds of CPU time)</P><P></P><P>I also tried Linlin's and Haikuo's with the 1 million obs ... both were about the same amount of time as above, 0.5 seconds</P><P></P><P>last, I tried KSharp's really elegant looking SQL code (that really is clever), stopping at 10,000 obs since even that takes about 11 seconds of CPU time</P></BODY></HTML> Thu, 30 Aug 2012 20:20:05 GMT https://communities.sas.com/t5/SAS-Procedures/how-to-program-like-this/m-p/107345#M29866 MikeZdeb 2012-08-30T20:20:05Z Re: how to program like this? https://communities.sas.com/t5/SAS-Procedures/how-to-program-like-this/m-p/107346#M29867 <HTML><HEAD></HEAD><BODY><P>Mike,</P><P></P><P>I like your new structure ... much easier to read than mine. </P><P></P><P>A quick question about testing Linlin's code ... it looks like you were able to use this syntax on a temporary array:</P><P></P><P>mean (of v(*))</P><P></P><P>I got an error trying that, but then I'm using 9.1.  I guess a later release must have added functions across temporary arrays.</P><P></P><P>Thanks for testing and posting.</P></BODY></HTML> Thu, 30 Aug 2012 20:48:28 GMT https://communities.sas.com/t5/SAS-Procedures/how-to-program-like-this/m-p/107346#M29867 Astounding 2012-08-30T20:48:28Z Re: how to program like this? https://communities.sas.com/t5/SAS-Procedures/how-to-program-like-this/m-p/107347#M29868 <HTML><HEAD></HEAD><BODY><P>hi ... correct, I just tried that in V9.1 and got ...</P><P></P><P>3    data _null_;</P><P>4    array v(5) _temporary_ (1 2 3 4 5);</P><P>5    mean = mean(of v(*));</P><P>ERROR: The ARRAYNAME<LI> specification requires a variable based array.</LI></P><P>6    put mean=;</P><P>7    run;</P><P></P><P>one reason to use that temporary array in Linlin's code is that such arrays are automatically retained</P><P></P><P>in V9.1, you can change that temporary array to a variable based array and add a RETAIN statement </P><P></P><P>then Linlin's code will work</P><P></P><P><SPAN style="font-family: 'courier new', courier;"><STRONG>data temp;</STRONG></SPAN></P><P><SPAN style="font-family: 'courier new', courier;"><STRONG>array v(0:4);</STRONG></SPAN></P><P><SPAN style="font-family: 'courier new', courier;"><STRONG>retain v;</STRONG></SPAN></P><P><SPAN style="font-family: 'courier new', courier;"><STRONG>set have;</STRONG></SPAN></P><P><SPAN style="font-family: 'courier new', courier;"><STRONG>v(mod(_n_,5))=var;</STRONG></SPAN></P><P><SPAN style="font-family: 'courier new', courier;"><STRONG>if _n_>=6 then do; mean_5=mean( of v(*)); output;end;</STRONG></SPAN></P><P><SPAN style="font-family: 'courier new', courier;"><STRONG>keep mean_5;</STRONG></SPAN></P><P><SPAN style="font-family: 'courier new', courier;"><STRONG>run;</STRONG></SPAN></P></BODY></HTML> Thu, 30 Aug 2012 21:18:31 GMT https://communities.sas.com/t5/SAS-Procedures/how-to-program-like-this/m-p/107347#M29868 MikeZdeb 2012-08-30T21:18:31Z Re: how to program like this? https://communities.sas.com/t5/SAS-Procedures/how-to-program-like-this/m-p/107348#M29869 <HTML><HEAD></HEAD><BODY><P>Thank you Mike and Astounding! I am happy<img id="smileyhappy" class="emoticon emoticon-smileyhappy" src="https://communities.sas.com/i/smilies/16x16_smiley-happy.png" alt="Smiley Happy" title="Smiley Happy" /> because my code is the shortest:smileysilly::smileylaugh:!</P><P></P><P>Linlin</P></BODY></HTML> Thu, 30 Aug 2012 21:48:38 GMT https://communities.sas.com/t5/SAS-Procedures/how-to-program-like-this/m-p/107348#M29869 Linlin 2012-08-30T21:48:38Z Re: how to program like this? https://communities.sas.com/t5/SAS-Procedures/how-to-program-like-this/m-p/107349#M29870 <HTML><HEAD></HEAD><BODY><P>Linlin,</P><P></P><P>Your comment reminds me of a smallest program challenge, on SAS-L, some years ago.  Take a look at:</P><P><A href="http://listserv.uga.edu/cgi-bin/wa?S2=sas-l&D=1&H=0&O=D&T=1&q=&s=Smallest+SAS+program&f=&a=2004&b=2004" title="http://listserv.uga.edu/cgi-bin/wa?S2=sas-l&D=1&H=0&O=D&T=1&q=&s=Smallest+SAS+program&f=&a=2004&b=2004">Search results -- SAS-L</title><style type="text/css"></style><head><title>LISTSERV at the University of Georgia</A></P></BODY></HTML> Thu, 30 Aug 2012 21:59:16 GMT https://communities.sas.com/t5/SAS-Procedures/how-to-program-like-this/m-p/107349#M29870 art297 2012-08-30T21:59:16Z Re: how to program like this? https://communities.sas.com/t5/SAS-Procedures/how-to-program-like-this/m-p/107350#M29871 <HTML><HEAD></HEAD><BODY><P>Art,</P><P>I have looked at the link. Too bad you did not win<img id="smileysad" class="emoticon emoticon-smileysad" src="https://communities.sas.com/i/smilies/16x16_smiley-sad.png" alt="Smiley Sad" title="Smiley Sad" />.</P></BODY></HTML> Thu, 30 Aug 2012 23:34:53 GMT https://communities.sas.com/t5/SAS-Procedures/how-to-program-like-this/m-p/107350#M29871 Linlin 2012-08-30T23:34:53Z Re: how to program like this? https://communities.sas.com/t5/SAS-Procedures/how-to-program-like-this/m-p/107351#M29872 <HTML><HEAD></HEAD><BODY><P>Linlin: ah, but I did (in a sense).  That, and about another gazillion similar posts, got me into the SAS-L Hall of Fame (see: <A href="http://www.sascommunity.org/wiki/SAS-L_BOF" title="http://www.sascommunity.org/wiki/SAS-L_BOF">http://www.sascommunity.org/wiki/SAS-L_BOF</A> )</P></BODY></HTML> Fri, 31 Aug 2012 00:00:20 GMT https://communities.sas.com/t5/SAS-Procedures/how-to-program-like-this/m-p/107351#M29872 art297 2012-08-31T00:00:20Z