About tc

tc · ‎11-19-2020

Yea, I'm getting spoiled by higher-resolution displays. 😀 I did try solid lines, which could be seen better, but I didn't like the way they bisected the state codes (also tried backlighting letters, but felt that varied letter heights too much for this particular chart). In retrospect, I guess I probably should have tried varying the thickness of the dotted line though. Oh well, there's always next time!

tc · ‎11-06-2020

Don't Miss the Gorilla in the Data! Had to do a quick scatter plot of the data given to students to analyze from Itai Yanai and Martin Lercher's A Hypothesis is a Liability, which is currently getting lots of Twitter love. *==> Fun w/SAS ODS Graphics: Don't Miss the Gorilla in the Data! Reference: A Hypothesis is a Liability, by Itai Yanai & Martin Lercher https://genomebiology.biomedcentral.com/articles/10.1186/s13059-020-02133-w Dropbox data files courtesy of the researchers; filename w url 'https://www.dropbox.com/s/685pkte3n3879mn/data9b_w.txt/?dl=1'; * Women's data; proc import datafile=w dbms=dlm out=w replace; delimiter='09'x; filename m url 'https://www.dropbox.com/s/r3wyn2ex20glsoa/data9b_m.txt/?dl=1'; * Men's data; proc import datafile=m dbms=dlm out=m replace; delimiter='09'x; proc sql; * Merge data together, assign category; create table StepsBMI as select 'WOMEN' as Category, w.* from w union all select 'MEN', m.* from m; proc sgplot aspect=1 noborder nowall; * Quick scatter plot to see what we've got!; styleattrs datacontrastcolors=(blue red); scatter x=steps y=bmi / group=category markerattrs=(symbol=circlefilled size=5pt); keylegend / location=inside position=topright down=2 noborder title=""; xaxis display=(noline noticks) grid gridattrs=(thickness=1pt color=darkgrey) valuesformat=comma9.; yaxis display=(noline noticks) grid gridattrs=(thickness=1pt color=darkgrey); label bmi="BODY MASS INDEX" steps="STEPS";

tc · ‎11-05-2020

Sure, it was just the 2020 state prediction data from CNN. To obtain the data, I did a "View Source" on the web page at https://www.cnn.com/election/2020/electoral-college-interactive-maps on 10/31, located the JSON data for the map, and simply copied-and-pasted the state entries into a text file, which I've uploaded (see attached file). Good luck!

ballardw · ‎09-07-2020

Shades of custom character sets from the days of 9-pin and similar impact printers. Fun.

Jay54 · ‎07-02-2020

Nice. And, if you use a GLOSS or SHEEN skin, you may also get the lighting effect on the shield. 🙂

tc · ‎05-25-2020

Memorial Day * Memorial Day 2020; data cross; * Create points for cross; color=.; * 10x10 base (no color); x=0; y=0; z=-.01; output; x=10; y=0; z=-.01; output; x=0; y=10; z=-.01; output; x=10; y=10; z=-.01; output; color=1; * 1=Light grey; x=4.5; y=4.5; z=0; output; * 3D cross vertices; x=5.5; y=4.5; z=0; output; x=4.5; y=5.5; z=0; output; x=5.5; y=5.5; z=0; output; x=4.5; y=5.5; z=10.5; output; x=5.5; y=5.5; z=10.5; output; x=4.5; y=7.5; z=10.5; output; x=5.5; y=7.5; z=10.5; output; x=4.5; y=7.5; z=12; output; x=5.5; y=7.5; z=12; output; x=4.5; y=5.5; z=12; output; x=5.5; y=5.5; z=12; output; x=4.5; y=5.5; z=17; output; x=5.5; y=5.5; z=17; output; x=4.5; y=4.5; z=17; output; x=5.5; y=4.5; z=17; output; x=4.5; y=4.5; z=12; output; x=5.5; y=4.5; z=12; output; x=4.5; y=2.5; z=12; output; x=5.5; y=2.5; z=12; output; x=4.5; y=2.5; z=10.5; output; x=5.5; y=2.5; z=10.5; output; x=4.5; y=4.5; z=10.5; output; x=5.5; y=4.5; z=10.5; output; x=4.5; y=4.5; z=0; output; x=5.5; y=4.5; z=0; output; x=4.5; y=5.5; z=0; output; x=5.5; y=5.5; z=0; output; proc template; * Use GTL to draw 3D cross; define statgraph cross; begingraph; layout overlay3d / cube=false xaxisopts=(display=none) yaxisopts=(display=none) zaxisopts=(display=none); surfaceplotparm x=x y=y z=z / surfacetype=fill colorresponse=color colormodel=(lightgrey); endlayout; endgraph; end; proc sgrender data=cross template=cross; run;

tc · ‎05-10-2020

You know, I forgot there was an ELLIPSEPARM statement for PROC SGPLOT! Gave it a quick shot, and it worked fine for generating a circle to replace the SCATTER plot, but I couldn't get a good shape for the petals. Also, I found that tweaking the equation for the petals yielded a more pleasing (to me at least!) shape for the petals: Take 2! * Fun With SAS ODS Graphics: Mother's Day Polar Flower References: blogs.sas.com/content/iml/2015/12/16/polar-rose.html matrixlab-examples.com/polar-plots.html thebestideasforkids.com/flower-template; data flower; /* Generate points for "flower" */ id=1; do theta = 0 to 2*constant("pi") by 0.01; r = abs(sin(3*theta))**(1/4); /* Six-petaled "flower" */ x = r*cos(theta); /* Convert to x/y */ y = r*sin(theta); output; end; x=.; y=.; x2=0; y2=0; mom='MOM'; output; /* Generate "pistil" */ /* Polygon/ellipse/text plots */ proc sgplot data=flower aspect=1 noborder noautolegend; polygon x=x y=y id=id / fill fillattrs=(color=cxb19cd9); ellipseparm semimajor=.35 semiminor=.35 / xorigin=0 yorigin=0 fillattrs=(color=cxfadadd) nooutline; text x=x2 y=y2 text=mom / textattrs=(color=white size=28pt color=purple); xaxis display=none min=-1 max=1 offsetmin=0 offsetmax=0; yaxis display=none min=-1 max=1 offsetmin=0 offsetmax=0;

tc · ‎04-20-2020

All fair points - sometimes I try to conserve graph "real estate" at the cost of explainability. And I guess this would be about as good a place as any for a "broken" axis!

tc · ‎03-15-2020

Right, you are, Rick! Modified above code/image (tossed in a radius line and titles, too).

Kurt_Bremser · ‎02-18-2020

@eajohnson wrote: Thanks so much @ChrisHemedinger. Do you ever consider matching your GF proceedings paper length to your session type? No. Last year I did a presentation on the Maxims of Maximally Efficient SAS Programmers; the paper contains all Maxims, while the presentation covered only 14 (and still ran for about 40 minutes, leaving enough time for some nice Q&A).

Ksharp · ‎02-09-2020

I like using symbolchar statement to draw a rectangle , and dropline statement to repalace block statement

ChrisHemedinger · ‎01-10-2020

Thanks to the community and especially @tc for this -- we adapted the fireworks to celebrate the 10,000 followers on the sassoftware Instagram handle. Props!

DanH_sas · ‎01-02-2020

Very cool 🙂

Ksharp · ‎12-21-2019

How about this one ? data have; group=1; do x=1 to 11 by 2; y=1;output; end; do x=2 to 10 by 2; y=3;output; end; do x=3 to 9 by 2; y=5;output; end; do x=4 to 8 by 2; y=7;output; end; do x=5 to 7 by 2; y=9;output; end; x=6;y=11;output; group=2; do x=1 to 11 by 2; y=2;output; end; do x=2 to 10 by 2; y=4;output; end; do x=3 to 9 by 2; y=6;output; end; do x=4 to 8 by 2; y=8;output; end; do x=5 to 7 by 2; y=10;output; end; x=6;y=12;output; run; data have; set have; if group=1 then y=y-0.38; if group=2 then y=y+0.38; run; ods graphics / attrpriority=none ; proc sgplot data=have noautolegend noborder aspect=1; styleattrs datasymbols=(triangledownfilled trianglefilled) datacontrastcolors=(cx00FF3E cx00B32C); scatter x=x y=y /group=group markerattrs=( size=50) ; xaxis display=none; yaxis display=none; run;

tomrvincent · ‎11-27-2019

I guess it doesn't work with Linux.

Online Status	Offline
Date Last Visited	a month ago