I have 10 data points representative of years (2003-2012). I would like to show the regression line for the first 8 points (2003-2010) and have the last two points stand alone. Is it possible to accomplish this with GPlot?
Easily done with SGPLOT. Move the last two points to new columns (x2, y2) and make them missing in the original columns (x, y). Now plot a regressioin plot using original columns (x, y). Overlay a scatter plot with new columns (x2, y2).
Great! I am going to try that. Thank you!
Well this was working great until I added the regression line. The regression line is stretching across all 10 points instead of stopping at 8.
Are you using gplot, or sgplot?
gplot
The built-in gplot interpol=rl regression line is going to always go all the way to the axis (I think).
If you just want the regression line to go to a certain point (and not all the way to the axis), then I think you'll need to calculate the points along the regression line (such as using Proc Reg), and then plot those points using interpol=join.
Or, just use SGPLOT.
data fit;
do x2=1 to 8;
y2=10+2*rannor(2); output;
end;
x2=.; y2=.;
do x=9 to 10;
y=10+2*rannor(2); output;
end;
run;
proc sgplot data=fit;
reg x=x2 y=y2 / name='r';
scatter x=x y=y;
keylegend 'r';
run;
Similar to what Sanjay was saying about sgplot, you could also do this with gplot.
See the 'Forecast Line' example, near the middle of this Line Plot chapter for an example:
Are you ready for the spotlight? We're accepting content ideas for SAS Innovate 2025 to be held May 6-9 in Orlando, FL. The call is open until September 25. Read more here about why you should contribute and what is in it for you!
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.