Help using Base SAS procedures

Regression testing in SAS

Accepted Solution Solved
Reply
Contributor
Posts: 60
Accepted Solution

Regression testing in SAS

Hi

I have recently bought the SAS program and ordered a book to help me unfortunately it is taking longer than expected to arrive, so please excuse me if this is a idiotic question.

I have a regression:

D = a +bx +e

For my data I have the dependent variable (D), along the explanatory variable (x). I aim to find the values for a and b. I would also like to find the r squared.

Suppose my dataset is called 'Accountingvar', D is the 'profit' and x is the 'sales price'.

What are the commands to perform this regression testing and R squared in SAS? Any help or further reading recommendations would be greatly appreciated


Accepted Solutions
Solution
‎04-26-2012 11:24 AM
Trusted Advisor
Posts: 2,115

Re: Regression testing in SAS

Posted in reply to spraynardz90

look at the documentation on support.sas.com for

PROC REG

PROC GLM

The syntax would be something like

PROC GLM DATA=<mydata>;

  MODEL d=x;

  RUN;

You may need to look at the options for the MODEL statement.

Doc Muhlbaier

Duke

View solution in original post


All Replies
Solution
‎04-26-2012 11:24 AM
Trusted Advisor
Posts: 2,115

Re: Regression testing in SAS

Posted in reply to spraynardz90

look at the documentation on support.sas.com for

PROC REG

PROC GLM

The syntax would be something like

PROC GLM DATA=<mydata>;

  MODEL d=x;

  RUN;

You may need to look at the options for the MODEL statement.

Doc Muhlbaier

Duke

Super User
Posts: 10,023

Re: Regression testing in SAS

Posted in reply to spraynardz90

Of course. Don't forget to check residual item to sure your model satisfied hypothesis .

Ksharp

Contributor
Posts: 60

Re: Regression testing in SAS

Thanks! With the REG Procedure results, what do I look at to check the residual item to ensure it satisfied my hypothesis? Do I want the value high or value low?

🔒 This topic is solved and locked.

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

Discussion stats
  • 3 replies
  • 248 views
  • 3 likes
  • 3 in conversation