| Date: | Mon, 1 Nov 2004 23:41:32 -0500 |
|---|
| Reply-To: | Rich.Ulrich@comcast.net |
|---|
| Sender: | "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU> |
|---|
| From: | Richard Ulrich <Rich.Ulrich@COMCAST.NET> |
|---|
| Subject: | Re: Standard error |
|---|
| Content-Type: | text/plain; charset=us-ascii |
|---|
On 1 Nov 2004 17:36:16 -0800, a320@hotmail.com (a320) wrote:
> Hello,
>
> Consider the regression Y = Bo + B1X + ... + BnXn
>
> SAS provides the standard error of B1 - SE(B1). If I multiply the B1
> by 0.1, is the SE of 0.1*B1 equal to 0.1*SE(B1)?
>
If I say that a regression shows that my grass grows
6 inches per month with a SE of 1.2, what is the meaning
of asking about 0.1 times 6?
Do you want to ask about change-of-scale?
Yes, 0.5 feet per month will have a SE of 0.1.
--
Rich Ulrich, wpilib@pitt.edu
http://www.pitt.edu/~wpilib/index.html
| 
