How can I put a equation in my regression graph ?

[josuerodrigues]

Hi. I'm working with doses of limestone and styding soil pH. My model is a quadratic polynomial and I did a scatterplot in PROC SGPLOT. But I want to know if I can put the equation inside of the graph. How can I do that? I know how to do in Excell, but in the SAS on Demand I don't know if it's possible. Below is the code line that I used. Thanks for the helpful.

proc sgplot data=mydata noautolegend;
title height=11pt color=black "soil pH vs limestone";
scatter Y=pH X=DOSE;
yaxis label="pH";
xaxis min=0 max=1.92 values=(0 0.48 0.96 1.44 1.92) label="Limestone (t/ha)";
REG Y=pH X=DOSE/degree=2;
keylegend / location=inside position=top across=5 down=3 titleattrs=(wight=bold size=9pt) valueattrs=(collor=black size=7pt);
run;

[Rick_SAS]

First, use PROC GLM to get the parameter estimates for the regression model:

data Have;
set sashelp.cars;
x = EngineSize;
y = MPG_City;
run;

proc glm data=Have;
model y = x x*x / solution;
quit;

Then you have two choices:

1. Display the regression coefficients in a table as name-value pairs

2. Display the regression line as an equation

For the first case (the table), here is an example:

proc sgplot data=Have noautolegend;
   reg Y=y X=x / degree=2;
   inset("Intercept"= "39.26"
         "x"   = "-8.65"
         "x^2" = "0.74") 
         / title="Regression Coefficients" opaque border;
run;

For the second case (an equation), here is an example:

proc sgplot data=Have noautolegend;
   reg Y=y X=x / degree=2;
   inset "y = 39.26 - 8.65*x + 0.74*x^2" / opaque border;
run;

If necessary, there are also ways to get an inline superscipt instead of x^2.