Ever wonder which cities are the most popular SASGF hosts? Well, wonder no more! Drawn from past, present, and future SASGF conference location info, here's a short program that creates a SAS ODS Graphics scatter plot showing SASGF 1976-2020 host cities together with markers for the years they've hosted a conference, the attendance for each year, and the total number of conferences hosted. The cities, which are ordered by the most recent year they hosted SASGF, include many one-hit wonders, as well as perennial favorites Orlando (6), San Francisco (4), Dallas (4), and Las Vegas (4). Believe it or not, this isn't the first time I've taken a crack at visualizing SASGF location data - check out Fit-to-a-T: Using SAS to Create a SAS Global Forum 2011 T-Shirt for a SAS/GRAPH (PROC GMAP) take on things!
CODE
* Fun w/SAS ODS Graphics: Scatter Plot of SASGF 1976-2020 Host Cities;
proc format; * Attendance, year formats;
picture attendf .='?' other='0,000';
picture yearf other='99';
data SASGFlocations; * Read past/present/future SASGF info;
infile "/folders/myfolders/SASGFconferences.txt" truncover;
input ConferenceName $30. / ConferenceDates $30. / City $30. / Attendees;
Year=input(trim(scan(ConferenceDates,-1)),4.);
proc sql; * Append # of conferences hosted;
create table SASGFlocationsFREQ as
select t1.city, put(attendees,attendf.) as AttendeesF, t1.year format=yearf., t2.FREQ
from SASGFlocations t1, (select city, count(*) as FREQ from SASGFlocations group by 1) t2
where t1.city=t2.city;
ods listing image_dpi=300 gpath='/folders/myfolders'; * Create 12"x7" scatter plot!;
ods graphics / reset=all width=12in height=7in imagefmt=GIF antialias imagename="SASGF_LOCATIONS";
proc sgplot data=SASGFlocationsFREQ;
title height=10pt "SAS GLOBAL FORUM HOST CITIES & ATTENDANCE (1976-2020)";
scatter x=year y=City / datalabel=attendeesF datalabelpos=right datalabelattrs=(size=7pt) markerattrs=(symbol=circlefilled) labelstrip;
xaxis grid display=(nolabel) valueattrs=(size=7pt) values=(1976 to 2020 by 1) offsetmin=.02 offsetmax=.02;
yaxis grid display=(nolabel) valueattrs=(size=8pt);
yaxistable FREQ / titleattrs=(size=8pt) stat=mean nolabel location=outside position=right;DATA
SAS Global Forum 2020 March 29 - April 1, 2020 Washington, DC ? SAS Global Forum 2019 April 28 - May 1, 2019 Dallas, TX ? SAS Global Forum 2018 April 8 - 11, 2018 Denver, CO ? SAS Global Forum 2017 April 2 - 5, 2017 Orlando, FL ? SAS Global Forum 2016 April 18 - 21, 2016 Las Vegas, NV 4917 SAS Global Forum 2015 April 26 - 29, 2015 Dallas, TX 4488 SAS Global Forum 2014 March 23 - 26, 2014 Washington, DC 4306 SAS Global Forum 2013 April 28 - May 1, 2013 San Francisco, CA 4118 SAS Global Forum 2012 April 22-25, 2012 Orlando, FL 3422 SAS Global Forum 2011 April 4-7, 2011 Las Vegas, NV 3408 SAS Global Forum 2010 April 11-14, 2010 Seattle, WA 2999 SAS Global Forum 2009 March 22-25, 2009 Washington, DC 3328 SAS Global Forum 2008 March 16-19, 2008 San Antonio, TX 3794 SAS Global Forum 2007 April 16-19, 2007 Orlando, FL 3655 SUGI 31 March 26-29, 2006 San Francisco, CA 4242 SUGI 30 April 10-13, 2005 Philadelphia, PA 3258 SUGI 29 May 9 - 12, 2004 Montreal, Quebec, CAN 2987 SUGI 28 March 30 - April 2, 2003 Seattle, WA 3036 SUGI 27 April 14-17, 2002 Orlando, FL 3676 SUGI 26 April 22-25, 2001 Long Beach, CA 3720 SUGI 25 April 9-12, 2000 Indianapolis, IN 3000 SUGI 24 April 11-14, 1999 Miami Beach, FL 3775 SUGI 23 March 22-25, 1998 Nashville, TN 3000 SUGI 22 March 16-19, 1997 San Diego, CA 3000 SUGI 21 March 10-13, 1996 Chicago, IL 3000 SUGI 20 April 2-5, 1995 Orlando, FL 2600 SUGI 19 April 10-13, 1994 Dallas, TX 2500 SUGI 18 May 9-12, 1993 New York, NY 3500 SUGI 17 April 12-15, 1992 Honolulu, HI 2500 SUGI 16 February 17-20, 1991 New Orleans, LA 3500 SUGI 15 April 1-4, 1990 Nashville, TN 3500 SUGI 14 April 9-12, 1989 San Francisco, CA 4000 SUGI 13 March 27-30, 1988 Orlando, FL 3500 SUGI 12 February 8-11, 1987 Dallas, TX 2800 SUGI 11 February 9-12, 1986 Atlanta, GA 2600 SUGI 10 March 10-13, 1985 Reno, NV 2500 SUGI '84 March 18-21, 1984 Hollywood Beach, FL 2092 SUGI '83 January 16-19, 1983 New Orleans, LA 1656 SUGI '82 February 14-17, 1982 San Francisco, CA 1322 SUGI '81 February 8-11, 1981 Lake Buena Vista, FL 1020 SUGI '80 February 18-20, 1980 San Antonio, TX 600 SUGI '79 January 29-31, 1979 Clearwater, FL 500 SUGI '78 January 30 - February 1, 1978 Las Vegas, NV 241 SUGI '77 January 3-5, 1977 New Orleans, LA 200 SAS.ONE '76 January 26-28, 1976 Kissimmee, FL 206
Good suggestion (and easy enough to change!). Looks good on $249 home laptop, be interesting to see how it looks on one of the 4K-resolution monitors at work. 
[click image to see full size version]
CODE
* Fun w/SAS ODS Graphics: Bubble Plot of SASGF 1976-2020 Host Cities;
proc format; * Attendance, year formats;
picture attendf 0='TBD' other='0,000';
picture yearf other='99';
data SASGFlocations; * Read past/present/future SASGF info;
infile "/folders/myfolders/SASGFconferences.txt" truncover;
input ConferenceName $30. / ConferenceDates $30. / City $30. / Attendees;
Year=input(trim(scan(ConferenceDates,-1)),4.);
if attendees=. then attendees=0;
proc sql; * Append # of conferences hosted;
create table SASGFlocationsFREQ as
select t1.city, attendees, put(attendees,attendf.) as AttendeesF, t1.year format=yearf., t2.FREQ
from SASGFlocations t1, (select city, count(*) as FREQ from SASGFlocations group by 1) t2
where t1.city=t2.city;
ods listing image_dpi=300 gpath='/folders/myfolders'; * Create 20"x14" bubble plot!;
ods graphics / reset=all width=20in height=14in imagefmt=GIF antialias imagename="SASGF_LOCATIONS";
proc sgplot data=SASGFlocationsFREQ;
title height=14pt "SAS GLOBAL FORUM 1976-2020 HOST CITIES AND NUMBER OF ATTENDEES";
bubble x=year y=City size=attendees /
colormodel=(deepskyblue darkblue) colorresponse=attendees bradiusmin=.125in datalabel=attendeesF
datalabelpos=center datalabelattrs=(size=7pt weight=bold color=white) dataskin=crisp;
gradlegend / notitle;
xaxis grid display=(nolabel) valueattrs=(size=7pt weight=bold) values=(1976 to 2020 by 1) offsetmin=.02 offsetmax=.02;
yaxis grid display=(nolabel) valueattrs=(size=9pt weight=bold);
yaxistable FREQ / valueattrs=(size=9pt weight=bold) stat=mean nolabel location=inside position=right;
Quick and dirty. Larger is more crowd, fainter is older.
data ATTENDANCE;
infile cards pad;
input NAME $32.
/ @', ' YEAR 4.
/ CITY $32.
/ ATTENDANCE 6.;
STATECODE=left(scan(CITY,2,','));
CITY=scan(CITY,1,',');
cards;
run;
proc sql;
create table COORDINATES as select *
from ATTENDANCE
left join
MAPS.USCITY
on ATTENDANCE.CITY = USCITY.CITY
and ATTENDANCE.STATECODE = USCITY.STATECODE
and USCITY.FEATYPE = 'city'
where ATTENDANCE.STATECODE not in ('HI' 'AK' 'PR' 'Quebec')
order by ATTENDANCE;
quit;
data ANNO;
set COORDINATES;
retain FUNCTION 'pie ' XSYS YSYS '2' ROTATE 360 WHEN 'A'
STYLE 'psolid' ;
COLOR=cat('Aff0000',20+2017-YEAR); %* Aff does alpha-transparency ;
SIZE=sqrt(ATTENDANCE/3.14)*.08;
run;
goptions device=png xpixels=600 ypixels=480;
pattern c=cxeeeeff;
ods html;
proc gmap data= MAPS.US map= MAPS.US anno=ANNO all;
where STATECODE not in ('HI' 'AK' 'PR' );
id STATE;
choro STATE / levels=1 nolegend;
run;
ods html close;
Adding an outer border allows better a view of the multiple occurrences.
proc sql;
create table COORDINATES as select *
from ATTENDANCE
left join
MAPS.USCITY
on ATTENDANCE.CITY = USCITY.CITY
and ATTENDANCE.STATECODE = USCITY.STATECODE
and USCITY.FEATYPE = 'city'
where ATTENDANCE.STATECODE not in ('HI' 'AK' 'PR' 'Quebec')
order by ATTENDANCE desc;
quit;
data ANNO;
set COORDINATES;
retain FUNCTION 'pie ' XSYS YSYS '2' ROTATE 360 WHEN 'A' ;
STYLE='psolid' ;
COLOR=cat('Aff0000',20+2017-YEAR); %* Aff for alpha-transparency ;
SIZE=sqrt(ATTENDANCE/3.14)*.1;
output;
STYLE='pempty' ;
COLOR='cx777777' ;
output;
run;
How to Concatenate Values
Learn how use the CAT functions in SAS to join values from multiple variables into a single value.
Find more tutorials on the SAS Users YouTube channel.
-
3056 views
-
21 likes
-
-
