I have used NLIN to model gas production curves using three different functions (Gompertz, Logistic, and LogLogistic). I am displaying the Raw Observations and the predicted curves on a graph. I am able to control colors of the lines but am not sure how to label them on the graph. The code I am using is shown below. It generates graphs that each contain 4 lines, as specified, but there is no legend.
Any help you could provide would be appreciated.
proc sort data=plot;by Treatment Replicate time;
ods graphics/height=6 in width=10 in noborder outputfmt=png ANTIALIASMAX=5600;
Title 'Gas Production';
proc sgpanel data=plot noautolegend;by Replicate;
label GasProd='Data' y_log_logistic='Log-logistic curve' y_logistic='Logistic curve' y_Gompertz='Gompertz curve'
ms_log='Log-logistic MS' ms='Logistic MS' ms_Gomp='Gompertz MS';
panelby Treatment/headerbackcolor=white novarname rows=1 columns=1;
series x=time y=GasProd/lineattrs=(color=gray pattern=1) group=Replicate;
series x=time y=y_log_logistic/lineattrs=(color=blue pattern=2) group=Replicate;
series x=time y=y_logistic/lineattrs=(color=red pattern=2) group=Replicate;
series x=time y=y_Gompertz/lineattrs=(color=green pattern=2) group=Replicate;
inset ms ms_log ms_Gomp/position=bottomright ;
If you remove the NOAUTOLEGEND option, does it give you something reasonable?
proc sgpanel data=plot noautolegend;by Replicate;
Here are some thoughts:
Hope this helps!
You also could try CURVELABEL= option to make your graph look pretty.
proc sgpanel data=sashelp.stocks NOAUTOLEGEND ;
panelby stock/onepanel columns=1;
series x=date y=close/curvelabel='close' curvelabelpos=max;
series x=date y=open/curvelabel='open' curvelabelpos=max;
run;
It's finally time to hack! Remember to visit the SAS Hacker's Hub regularly for news and updates.
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.
Ready to level-up your skills? Choose your own adventure.