Message-ID: <15852484.post@talk.nabble.com>
Date: 2008-03-05T15:16:51Z
From: David Hewitt
Subject: non-linear correlation
In-Reply-To: <E7163EE5889C364389C921C0DE04152401FD01@lu-mail-san.dfu.local>
> Which effect sizes (parametric or not) could I use in order to estimate
> the amount of non-linear correlation between 2 variables?
>
> Is it possible to correct for auto-correlation within the correlated
> times series?
>
I think the starting point is to develop a model, even conceptual, of how
you expect them to be related. If you've already determined that they are
related, and in a way that's nonlinear, then you can probably come up with a
model. "Effect size" will be dependent on the model, but in a nonlinear
model it won't be as simple as "y goes as x goes".
I can't see what you mean concerning the autocorrelation, but if this is a
time series problem your first question is then a mystery to me. You might
do well to read the posting guide for the mailing list and provide an
example.
-----
David Hewitt
Virginia Institute of Marine Science
http://www.vims.edu/fish/students/dhewitt/
--
View this message in context: http://www.nabble.com/non-linear-correlation-tp15848891p15852484.html
Sent from the R help mailing list archive at Nabble.com.