Data visualization with SAS programming

Proc template Regression Plot with No intercept

Accepted Solution Solved
Reply
Contributor
Posts: 46
Accepted Solution

Proc template Regression Plot with No intercept

[ Edited ]

Is there any option in proc template to plot regression plot with no intercept ?

 

 

 regressionplot x =  test1 y = test2 / name="fitline" legendlabel="Regression line";


Accepted Solutions
Solution
‎03-12-2018 02:59 PM
SAS Super FREQ
Posts: 496

Re: Proc template Regression Plot with No intercept

Here is an example with SGPLOT.  The NOINT option forces the line through 0,0.  The output statement stores the predicted values in the variable p.

 

proc reg data=sashelp.class;
   model weight = height / noint;
   output out=p p=p;
quit;

proc sort data=p; by height; run;

proc sgplot data=p;
   scatter y=weight x=height;
   series y=p x=height;
run;

View solution in original post


All Replies
SAS Super FREQ
Posts: 496

Re: Proc template Regression Plot with No intercept

No.  You can run PROC REG and output the predicted values and then plot them using a series plot.

Contributor
Posts: 46

Re: Proc template Regression Plot with No intercept

Posted in reply to WarrenKuhfeld

Haven't done regression line without using regression plot. Which parameters/predicated values from proc reg should be use to plot this ?

Solution
‎03-12-2018 02:59 PM
SAS Super FREQ
Posts: 496

Re: Proc template Regression Plot with No intercept

Here is an example with SGPLOT.  The NOINT option forces the line through 0,0.  The output statement stores the predicted values in the variable p.

 

proc reg data=sashelp.class;
   model weight = height / noint;
   output out=p p=p;
quit;

proc sort data=p; by height; run;

proc sgplot data=p;
   scatter y=weight x=height;
   series y=p x=height;
run;
Contributor
Posts: 46

Re: Proc template Regression Plot with No intercept

Posted in reply to WarrenKuhfeld

Thank you so much. This is very helpful. 

☑ This topic is solved.

Need further help from the community? Please ask a new question.

Discussion stats
  • 4 replies
  • 131 views
  • 0 likes
  • 2 in conversation