topic Re: Using PROC REG to write prediction equations in SAS Programming

Using PROC REG to write prediction equations

JUMMY — Wed, 30 Jan 2019 04:15:38 GMT

I have a data of this form:

data brand;
	input preference moisturecontent sweetness @@;
	datalines;
	64 4 2 73 4 4 61 4 2 76 4 4 72 6 2 80 6 4 
	71 6 2 83 6 4 83 8 2 89 8 4 86 8 2 93 8 4
	88 10 2 95 10 4 94 10 2 100 10 4
	;
run;
proc print;
run;


proc reg data=brand;
model preference=moisturecontent sweetness;
run;

So I used PROC REG to write out the prediction equation. How to I write a separate prediction equation between brand liking and sweetness when moisture content equals to 4 and when moisture content equals to 10?

Re: Using PROC REG to write prediction equations

Reeza — Wed, 30 Jan 2019 04:12:54 GMT

Your code isn't correct. you can't have 'BY MAKE' since you have no make variable.

It's not quite clear what you question is, can you please clarify in detail?

Re: Using PROC REG to write prediction equations

JUMMY — Wed, 30 Jan 2019 04:19:15 GMT

@Reeza, I apologize for the mistake. I have corrected it. PROC REG gives the multiple regression line for the data. But I want to come up with two different equations for preference and sweetness when moisture content equals to 4 and when moisture content equals to 10.

Re: Using PROC REG to write prediction equations

Reeza — Wed, 30 Jan 2019 05:19:43 GMT

@JUMMY wrote:

@Reeza, I apologize for the mistake. I have corrected it. PROC REG gives the multiple regression line for the data. But I want to come up with two different equations for preference and sweetness when moisture content equals to 4 and when moisture content equals to 10.

Put that variable in the BY statement too. The BY statement can take multiple variables and will do it by nested groups.

By make sweetness;