----- Forwarded message from
[email protected] -----
Date: Tue, 26 Jun 2012
09:45:21 -0700
From: [email protected]
Reply-To: [email protected]
Subject: Re: problem
with RMA regression
To: [email protected]
----- Forwarded message from
Robert Corruccini
Date: Tue, 26 Jun 2012
08:16:45 -0400
From: Robert Corruccini
Reply-To: Robert Corruccini
Subject: Re: problem
with RMA regression
To:
[email protected]
On Mon, Jun 25, 2012 at 5:28
AM, <[email protected]>
wrote:
----- Forwarded message from mahdi rabie -----
Date: Mon, 18 Jun 2012 02:41:32 -0400
From: mahdi rabie
Reply-To: mahdi rabie
Subject: problem with RMA regression
To: "[email protected]"Dear AllI have a problem with RMA regression (Reduced Major Axis regression) and would be very grateful if somebody help me.I'm studying on allometry of somatic and sexual traits of two Noctuid moths. I want to calculate slope B of regression line for log transformed data for calculation of kind of allometry relationship (isometry or negative allometry or positive allometry ).I first calculated it by OLS regression and found negative allometry for some sexual traits and isometry and negative allometry for somatics. Again, I calculated B by method RMA by using RMA Software. I found isometry for often somatic traits and positive for some of them and positive allometry for sexual traits ! It is strange results !
I believe RMA
slopes are always positive -- the RMA slope equals OSA slope divided by r and a
negative OSA slope is always (mathematical necessity) accompanied by a negative
correlation ....
--
Robert S. Corruccini
Professor Emeritus and Distinguished Scholar
Department of Anthropology
Southern Illinois University
Carbondale, IL 62901-4502
----- End forwarded message -----
----- End forwarded message
-----
