s-news
[Top] [All Lists]

Re: Regression question

To: "'Chushu Gu'" <chushugu@hotmail.com>, s-news@lists.biostat.wustl.edu
Subject: Re: Regression question
From: "Liaw, Andy" <andy_liaw@merck.com>
Date: Wed, 28 Jan 2004 16:42:22 -0500
I'm not sure I understand exactly what you are looking for, but Prof. Fox's `car' library (http://socserv.socsci.mcmaster.ca/jfox/Books/Companion/car.html) has ceres.plot() and cr.plot(), which may do what you want.
 
HTH,
Andy
-----Original Message-----
From: s-news-owner@lists.biostat.wustl.edu [mailto:s-news-owner@lists.biostat.wustl.edu] On Behalf Of Chushu Gu
Sent: Wednesday, January 28, 2004 4:29 PM
To: s-news@lists.biostat.wustl.edu
Subject: Re: [S] Regression question

Thanks Adrienne Kandel,
 
I think my question is different.
 
To check collinearity, I do regression analysis on
x1 = a + b1 x2 + b2 x3 + e
x2 = ...
x3 = ...
 
What is partial regression line? For example, Is it a line Y against x1 when x2 x3 fixed
What is the use of partial regressional line, I am confused. Is it for detecting collinearity? (I just want to check linearity)
I got 200 obs on x1 x2 x3 Y. Since I can not get enough obs when x2 x3 fixed (Actually no same x2 x3), so how to plot this line.
As your explanation, this line seems unnecessary since I already use one method to detect collinearity between each variable of them.
I don't know how to plot partial regression line, but I plot a line Y against x1 (no fixed x2 x3). It has no pattern (I can not see a line, somewhere condensed,
 somewhere loose). I don't know how to check linearity (There is no line). I did regression on Y based on x1, R square is only 0.05. What should I do?  And I did regression on Y based on all three x, R square is only 0.2. Is it a failed regression analysis?
 
 
Regards,
 
Chushu Gu
 

------------------------------------------------------------------------------
Notice: This e-mail message, together with any attachments, contains information of Merck & Co., Inc. (One Merck Drive, Whitehouse Station, New Jersey, USA 08889), and/or its affiliates (which may be known outside the United States as Merck Frosst, Merck Sharp & Dohme or MSD and in Japan as Banyu) that may be confidential, proprietary copyrighted and/or legally privileged. It is intended solely for the use of the individual or entity named on this message. If you are not the intended recipient, and have received this message in error, please notify us immediately by reply e-mail and then delete it from your system.
------------------------------------------------------------------------------

<Prev in Thread] Current Thread [Next in Thread>