hi ... have you tried just using a format , for example the dates in SASHELP.AIR have year and month but you can plot by year on the x-axis ... goptions reset=all; proc gplot data=sashelp.air; plot air*date; format date year.; run; quit; ... View more

hi ... if you know something about the values of the BY-variable, you can use a WHERE statement for example ... data air; set sashelp.air; year = year(date); run; goptions reset=all; proc gplot data=air; plot air*date; by year; where year between 1950 and 1952; run; quit; if you want to limit the output by the number of plots, you could try something like this ... ods listing close; proc sql number; ods output sql_results=temp; select distinct year from air; ods output close; select year into :years  separated by ',' from temp where row le 5; quit; ods listing; * generates 5 plots;  proc gplot data=air; plot air*date; by year; where year in (&years); run; quit; ... View more

hi ... same idea as Art, but no IF-THEN ... data x; input id score @@; datalines; 100 10 100 13 100 16 100 19 101 13 101 14 101 15 101 16 103 20 103 22 103 24 103 26 ; data x; set x; by id; time + (-first.id * time) + 1; run; ... View more

Hi ... if anyone would like to do some reading (from 9 years ago with most of the same points brought up in the discussion) ... "Danger:  MERGE Ahead! Warning:  BY Variable with Multiple Lengths!" by Bob Virgile http://www.nesug.org/proceedings/nesug03/at/at005.pdf ... View more

hi ... some other ideas ... #1 this also uses the ZIPCITYDISTANCE function ... basically Robert's ideas in one use of PROC SQL, though I took out the condition  where usa1.zip^=metro_rank.metro_zip; since I think it'd be OK if the closest USA zip to a METRO zip was the same zip, yes/no? the first part cleans up USA1 it replaces spaces in ADDZIP with asterisks then creates a numeric variable UZIP using the 1st five characters of ADDZIP (you get some LOG messages about INVALID strings) it also checks to see if UZIP is in SASHELP.ZIPCODE since ZIPCITYDISTANCE uses that data set the second part grabs the top 50 observations in METRO_RANK the third part computes the distance and grabs the minimum distance between a TIC and a top 50 CITY you get some LOG messages about INVALID  strings proc sql; create view usa as select unique tic, input(translate(addzip,'*',' '), 5.) as uzip format=z5. from z.usa1, sashelp.zipcode (keep=zip) as zc having uzip eq zc.zip; create view metro as select city, sum_rank, input(zipcode, 5.) as mzip format=z5. from z.metro_rank (where=(sum_rank between 1 and 50)); create table pairs as select *, zipcitydistance(uzip, mzip) as dist from usa, metro group tic having dist eq min(dist); quit; #2  this would be too big a problem to for the SAS and Google Maps drive distance/time (not really intended for 500K+ calls to Google Maps) #3  the code in the original SGF paper no longer works ... a link to the most recent code canbe found at ... http://www.sascommunity.org/wiki/User:Msz03 (but remember, your problem is TOO BIG) #4  Darrell's comment of "... Google Maps terms of use ..." ... I looked (again) at ... http://www.google.com/help/terms_maps.html and see the two possible sticking points as how one would interpret ... (a) access or use the Products or any Content through any technology or means other than those provided in the Products, or through other explicitly authorized means Google may designate (such as through the Google Maps/Google Earth APIs); and (e) use the Products in a manner that gives you or any other person access to mass downloads or bulk feeds of any Content, including but not limited to numerical latitude or longitude coordinates, imagery, and visible map data; I see at ... http://blog.programmableweb.com/2011/10/27/google-maps-usage-fees-how-many-developers-will-have-to-pay/http://code.google.com/apis/maps/faq.html#usagelimits that folks using Google APIs for map access have a 25,000 free maps/day, after that there's a charge #5  take a look at ... http://www.tandfonline.com/doi/abs/10.1080/00330124.2011.583586#preview Boscoe FP, Henry KA, Zdeb MS. A Nationwide Comparison of Driving Distance Versus Straight Line Distance to Hospitals. Professional Geographer 2012 the conclusion in that paper is that in most instances, the added precision of driving distance over straight line distance isn't worth the effort ... View more

hi ... the big problem with using annotate to modify a star chart is that the data coordinate system cannot be used, so you'd have to figure out the locations of anything that has to be tweaked based on percentages of output area and that'll vary from chart to chart so, Art's overlay solution is a lot easier than annotate (but pretty limited as to what you can accomplish) ... View more

hi ... if you don't want to set an initial value, you can skip the RETAIN statement since a SUM statement (K+M) implies retain and starts with an initial value of 0 if you want to set an initial value ... RETAIN K 2.4; ps http://support.sas.com/documentation/cdl/en/lrdict/64316/HTML/default/viewer.htm#a000289454.htm ... View more