[R-sig-ME] LME and nonlinearity?
John.Maindonald at anu.edu.au
Fri Apr 10 01:19:09 CEST 2009
These various pieces of advice should come with a warning: Beware of
all generalized advice. It may miss key issues that are obvious only
once one has gained some sense of the scientific background, seen the
data, and done some preliminary analysis.
John Maindonald email: john.maindonald at anu.edu.au
phone : +61 2 (6125)3473 fax : +61 2(6125)5549
Centre for Mathematics & Its Applications, Room 1194,
John Dedman Mathematical Sciences Building (Building 27)
Australian National University, Canberra ACT 0200.
On 10/04/2009, at 9:08 AM, Samer Mouksassi wrote:
> Another possible way to diagnose the need of a non-linear model is
> to use a spline then replace it with more meaningful parametric
> model according to its shape.
> -----Original Message-----
> From: r-sig-mixed-models-bounces at r-project.org [mailto:r-sig-mixed-models-bounces at r-project.org
> ] On Behalf Of John Maindonald
> Sent: 2009-04-09 18:51
> To: Balázs Lestár
> Cc: r-sig-mixed-models at r-project.org
> Subject: Re: [R-sig-ME] LME and nonlinearity?
> Additional to the comments below: Think/check also whether
> transformation of one or more of the variables (log transformation?)
> makes the relationship more nearly linear.
> On 10/04/2009, at 7:13 AM, Balázs Lestár wrote:
>> Dear All,
>> I have a mixed model (LME), but one of my explanatory variables is
>> not linearly related to the dependent variable.
>> 1.) Somebody told me, to make a 2 or 3 level factor from the
>> continuous variable. (I wouldn't prefer this)
> In general, this makes poor use of the information in the data. You
> lose power.
>> 2.) I saw in some statistical books that in these cases, I have
>> to use in the model the quadratic term of the variable. (but the AIC
>> is much greater than with the factorized variable)
>> Is that possible, to use a poly() function in the lme? (this model
>> seems to be the best, based on AIC).
>> I'm a bit confused, 'cause the LME supposes linear relation between
>> variables. Isn't it right?
> Linear models are linear in the parameters. They can model highly
> nonlinear effects.
>> 3.) Do I need a non-linear model?
> Only if you need a model that is non-linear in the parameters.
> Without checking out your data and model, one cannot say.
> John Maindonald email: john.maindonald at anu.edu.au
> phone : +61 2 (6125)3473 fax : +61 2(6125)5549
> Centre for Mathematics & Its Applications, Room 1194,
> John Dedman Mathematical Sciences Building (Building 27)
> Australian National University, Canberra ACT 0200.
>> Which solution is the best?
>> R-sig-mixed-models at r-project.org mailing list
> R-sig-mixed-models at r-project.org mailing list
More information about the R-sig-mixed-models