Hi
Below sample data
visit dosem avalue
1 50 8.66
2 50 6.33
1 100 6.88
2 100 8.33
1 200 76.22
2 200 65.22
1 400 55.88
2 400 88.77
1 50 77.22
2 50 54.3
3 50 54
4 50 45
5 50 0.003
6 50 .001
I am trying to generate a plot whi ch will have all avalues as scatter points by each visitn
like dose*avalue=visit
i want to represent median line in the plot
also i want a regression ln(dose) = ln(avalue)
then (intercept*x**slope) is equation
VISIT variable decode
I calculated median and regression estimates values kept in one dataset
1-single dose scatter plot
2-multiple dose scatter plot
3.median of single dose
4.median of multiple dose
5.regression value of single dose
6.multiple dose regrssion value
then I used below procedure
symbol1 color=black value=circle height=1.5 mode=include INTERPOL=none;
symbol2 color=grey value=trianglefilled height=1.5 mode=include INTERPOL=none;
symbol3 color=blue w = 3 height=2 mode=include INTERPOL=hiloct;
symbol4 color=green w = 3 height=2 mode=include INTERPOL=hiloct ;
/* Define the symbol characteristics for the regression line */
symbol5 interpol=rl value=none color=black;
*symbol5 interpol=join value=dot color=black l=1;
*symbol6 interpol=join value=circle color=black l=2;
symbol6 interpol=rl value=none color=grey l=2;
proc sort data = x1;
by avisitn dose ;
run;
proc gplot data = x1;
plot avalue * dose = avisitn/ vaxis=axis1 haxis=axis2 noframe nolegend;
run;
quit;
My question my regression line not coming staright and also
I want both x,y scales in log base and custom tick marks
The data i gave example so values may not look correct
appreciate help
thank u
Hi @valli
If you want to overlay scatter, median line and regression line on one plot. There are two options:
1. PROC GPLOT, use PLOT, PLOT2 statements and ANNOTATION for three tasks. ex. you can use ANNOTATION for scatter, PLOT for median line and PLOT2 for regression line with SYMBOL statement to define INTERPOL=R<type><0><CLM | CLI<50...99>>. (http://www.okstate.edu/sas/v8/sashtml/gref/zbolchap.htm)
2. SGPLOT or SGSCATTER depend on the output you want to display.
Hope it will help.
Please submit usable data:
Your variable names are inconsistent, axis1 and axis2 statements are not provided.
Regarding the regression, did you see this message that your code generates? :
WARNING: Fewer than three node values for at least one observation in HILOC interpolation. Mean value used for tick placement.
Hi @valli
If you want to overlay scatter, median line and regression line on one plot. There are two options:
1. PROC GPLOT, use PLOT, PLOT2 statements and ANNOTATION for three tasks. ex. you can use ANNOTATION for scatter, PLOT for median line and PLOT2 for regression line with SYMBOL statement to define INTERPOL=R<type><0><CLM | CLI<50...99>>. (http://www.okstate.edu/sas/v8/sashtml/gref/zbolchap.htm)
2. SGPLOT or SGSCATTER depend on the output you want to display.
Hope it will help.
SAS Innovate 2025 is scheduled for May 6-9 in Orlando, FL. Sign up to be first to learn about the agenda and registration!
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.