Plot variables as percentage of another variable

[riccardo88]

Hallo everybody,

I´m plotting my data and I would like to show 4 variable as percentage of another variable.

My data:

My Code:

proc sgplot data=Transposed_average_hz;
scatter x=frequency y=col1;
scatter x=frequency y=col2;
scatter x=frequency y=col3;
scatter x=frequency y=col4;
scatter x=frequency y=col5;
run;

What I would like to have (sort of)...

Thank you very much for your help!

[PeterClemmensen]

What four variables do your want to plot? And what variable should they represent percentages of?

[riccardo88]

COL 2,3,4,5 as percentage of COL1.

COL1 should be a straight line in the graphic.

[PeterClemmensen]

Ok. Can you post your data in a usable form (Not a picture)?

[RW9]

Looks like Excel thinking again.  Get your data looking like:
COL    PCENT   FREQUENCY

1         ...             ...

....

Then use:

proc sgplot data=Transposed_average_hz;
  series x=frequency y=pcent / group=col;
run;

Note, as you said you want a line, then series is most likely. 

[Ksharp]

Your code look good. add option "curvelabel"

proc sgplot data=Transposed_average_hz;
scatter x=frequency y=col1/curvelabel ;
scatter x=frequency y=col2/curvelabel;
scatter x=frequency y=col3/curvelabel;
scatter x=frequency y=col4/curvelabel;
scatter x=frequency y=col5/curvelabel;
run;

[Rick_SAS]

Sort by the X variable and use the SERIES statement:

proc sort data=Transposed_average_hz;
by frequency;
run;

proc sgplot data=Transposed_average_hz;
series x=frequency y=col1 / curvelabel;
series x=frequency y=col2 / curvelabel;
series x=frequency y=col3 / curvelabel;
series x=frequency y=col4 / curvelabel;
series x=frequency y=col5 / curvelabel;
run;