[R-sig-ME] Fwd: Extremely High Slope/Intercept Correlation

Viechtbauer Wolfgang (STAT) wolfgang.viechtbauer at maastrichtuniversity.nl
Mon Dec 2 14:04:55 CET 2013


Ah okay, I misunderstood what 'Male' stood for.

To figure out what is going on, you could fit lm(Y ~ X) for each male separately (e.g., with the lmList function) and then examine a plot of the intercepts versus the slopes.

Best,
Wolfgang

> -----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 AvianResearchDivision
> Sent: Monday, December 02, 2013 13:43
> To: r-sig-mixed-models at r-project.org
> Subject: [R-sig-ME] Fwd: Extremely High Slope/Intercept Correlation
> 
> ---------- Forwarded message ----------
> From: AvianResearchDivision <segerfan83 at gmail.com>
> Date: Mon, Dec 2, 2013 at 7:42 AM
> Subject: Re: [R-sig-ME] Extremely High Slope/Intercept Correlation
> To: "Viechtbauer Wolfgang (STAT)" <
> wolfgang.viechtbauer at maastrichtuniversity.nl>
> 
> 
> Male is not dichotomous, it has 59 levels for 59 different males.  I run
> this same model, but with a different dependent variables and I get much
> lower correlations.  Can someone please explain what "This can be caused
> by
> having more complexity than the data allows. Note that the variance of the
> random slope is small." means?
> 
> Thank you,
> Jacob
> 
> 
> On Mon, Dec 2, 2013 at 4:35 AM, Viechtbauer Wolfgang (STAT) <
> wolfgang.viechtbauer at maastrichtuniversity.nl> wrote:
> 
> > I assume 'Male' is dichotomous, so it has just two levels. Not likely
> that
> > one can estimate any variance components based on that grouping variable
> > with any degree of precision.
> >
> > > -----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 ONKELINX, Thierry
> > > Sent: Monday, December 02, 2013 10:20
> > > To: AvianResearchDivision; r-sig-mixed-models at r-project.org
> > > Subject: Re: [R-sig-ME] Extremely High Slope/Intercept Correlation
> > >
> > > This can be caused by having more complexity than the data allows.
> Note
> > > that the variance of the random slope is small.
> > >
> > > -----Oorspronkelijk bericht-----
> > > Van: r-sig-mixed-models-bounces at r-project.org [mailto:
> > r-sig-mixed-models-
> > > bounces at r-project.org] Namens AvianResearchDivision
> > > Verzonden: maandag 2 december 2013 2:05
> > > Aan: r-sig-mixed-models at r-project.org
> > > Onderwerp: [R-sig-ME] Extremely High Slope/Intercept Correlation
> > >
> > > Hi all,
> > >
> > > I came across extremely high slope/intercept correlation (-1.00) when
> > > running the following model:
> > >
> > > model<-lmer(X~Y+(Y|Male))
> > >
> > > summary(model)
> > >
> > > Random effects:
> > >
> > >  Groups             Name               Variance          Std.Dev.
> > > Corr
> > >
> > >
> > >  Male                (Intercept)          1.562e-01        0.3952284
> > >
> > >                         Y                       7.581e-07
> 0.0008707
> > > -1.00
> > >
> > >                         Residual            2.661e-01        0.5158953
> > >
> > > Number of obs: 879, groups: Male, 59
> > >
> > >
> > > What is causing this?  This high of a correlation does not seem
> possible,
> > > especially in ecology.  Any suggestions moving forward?
> > >
> > >
> > > Thank you,
> > >
> > > Jacob
> > >
> > >         [[alternative HTML version deleted]]
> > >
> > > _______________________________________________
> > > R-sig-mixed-models at r-project.org mailing list
> > > https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models
> > > * * * * * * * * * * * * * D I S C L A I M E R * * * * * * * * * * * *
> *
> > > Dit bericht en eventuele bijlagen geven enkel de visie van de
> schrijver
> > > weer en binden het INBO onder geen enkel beding, zolang dit bericht
> niet
> > > bevestigd is door een geldig ondertekend document.
> > > The views expressed in this message and any annex are purely those of
> the
> > > writer and may not be regarded as stating an official position of
> INBO,
> > as
> > > long as the message is not confirmed by a duly signed document.
> > >
> > > _______________________________________________
> > > R-sig-mixed-models at r-project.org mailing list
> > > https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models
> >
> > _______________________________________________
> > R-sig-mixed-models at r-project.org mailing list
> > https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models
> >
> 
> 	[[alternative HTML version deleted]]
> 
> _______________________________________________
> R-sig-mixed-models at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models



More information about the R-sig-mixed-models mailing list