[R-sig-ME] correlated samples

Highland Statistics Ltd highstat at highstat.com
Wed Jul 13 10:14:37 CEST 2016


Joaquin,

I don;t know where you get the first Vi from. My final pdf shows an Y_i, 
as you can see in the snapshot below.





However....downloading the pdf seems that the online version of the book 
does have a V_i. I guess the editing company of Springer messed up the 
text (or I didn't double check their corrections good enough).

Thierry is right in stating that the equations are for a linear mixed 
effects model with a random intercept. If there are two random 
intercepts then the equations change (See Chapter 20 in the same book).

I did look at similar expressions for the intraclass correlation for (1 
way nested and 2-way nested) Poisson GLMM. They are given in our 2012 
book. You can also find them in Nzoufras (2009). I tried to derive them 
for ZIP models, but that was quite difficult. The following two papers 
do simulation to calculate the intra-class correlation for Bernoulli GLMMs:

Goldstein H, Browne W, Rasbash J (2002) Partitioning Variation in 
Multilevel Models.
Understanding Statistics 1: 223–231.

Browne WJ, Subramanian SV, Jones K (2005) Variance partitioning in 
multilevel logistic models that
exhibit overdispersion. Journal Royal Statistical Society. Series A 168: 
599–613.

I'm sure there will be some more recent material on intraclass 
correlations for complicated GLMMs.

Alain





On 13/07/2016 17:50, Thierry Onkelinx wrote:
> Dear Joaquin,
>
> You can derive eqn. 5.13 from the assumptions in eqn. 5.8, which 
> assumes that Yi follows a normal distribution.
>
> Note that the example is use in a chapter which introduces **linear** 
> mixed models. Hence a normal distribution is assumed for the sake of 
> simplicity.
>
> Best regards,
>
> Thierry
>
> ir. Thierry Onkelinx
> Instituut voor natuur- en bosonderzoek / Research Institute for Nature 
> and Forest
> team Biometrie & Kwaliteitszorg / team Biometrics & Quality Assurance
> Kliniekstraat 25
> 1070 Anderlecht
> Belgium
>
> To call in the statistician after the experiment is done may be no 
> more than asking him to perform a post-mortem examination: he may be 
> able to say what the experiment died of. ~ Sir Ronald Aylmer Fisher
> The plural of anecdote is not data. ~ Roger Brinner
> The combination of some data and an aching desire for an answer does 
> not ensure that a reasonable answer can be extracted from a given body 
> of data. ~ John Tukey
>
> 2016-07-12 19:42 GMT+02:00 Joaquín Aldabe <joaquin.aldabe at gmail.com 
> <mailto:joaquin.aldabe at gmail.com>>:
>
>     By the way..I don´t see why it is relatively easy to show that Yi
>     is normally distributed..(in fact, the Y variable is count so it
>     should be poisson). This is in section 5.4 (p. 112) of Zuur et al
>     (2009).
>     Thanks,
>     joaquín.
>
>     2016-07-12 10:52 GMT-03:00 Thierry Onkelinx
>     <thierry.onkelinx at inbo.be <mailto:thierry.onkelinx at inbo.be>>:
>
>         Dear Joaquin,
>
>         Yes. I think you have found a typo. The prior to equation 5.13
>         should read "It is relatively easy to show that $Y_i$ is
>         normally distributed ..."
>
>         Best regards,
>
>         ir. Thierry Onkelinx
>         Instituut voor natuur- en bosonderzoek / Research Institute
>         for Nature and Forest
>         team Biometrie & Kwaliteitszorg / team Biometrics & Quality
>         Assurance
>         Kliniekstraat 25
>         1070 Anderlecht
>         Belgium
>
>         To call in the statistician after the experiment is done may
>         be no more than asking him to perform a post-mortem
>         examination: he may be able to say what the experiment died
>         of. ~ Sir Ronald Aylmer Fisher
>         The plural of anecdote is not data. ~ Roger Brinner
>         The combination of some data and an aching desire for an
>         answer does not ensure that a reasonable answer can be
>         extracted from a given body of data. ~ John Tukey
>
>         2016-07-11 21:36 GMT+02:00 Joaquín Aldabe
>         <joaquin.aldabe at gmail.com <mailto:joaquin.aldabe at gmail.com>>:
>
>                 Sorrry, I hadn´t copied the whole group. Here it goes:
>                 Thanks Thierry, I´ve been there a couple of times.
>                 I´ll try again. I think there is a mistake in page 112
>                 when says: It is relatively easy to show that Vi is
>                 normally distributed with mean X i × β and variance V
>                 i in mathematical notation...
>
>                 It should say Yi instead of Vi, right?
>
>                 Cheers,
>                 Joaquín .
>
>                 2016-07-08 9:02 GMT-03:00 Thierry Onkelinx
>                 <thierry.onkelinx at inbo.be
>                 <mailto:thierry.onkelinx at inbo.be>>:
>
>                     Dear Joaquin,
>
>                     I'd recommend Zuur et al 2009, section 5.4
>
>                     Best regards,
>
>                     ir. Thierry Onkelinx
>                     Instituut voor natuur- en bosonderzoek / Research
>                     Institute for Nature and Forest
>                     team Biometrie & Kwaliteitszorg / team Biometrics
>                     & Quality Assurance
>                     Kliniekstraat 25
>                     1070 Anderlecht
>                     Belgium
>
>                     To call in the statistician after the experiment
>                     is done may be no more than asking him to perform
>                     a post-mortem examination: he may be able to say
>                     what the experiment died of. ~ Sir Ronald Aylmer
>                     Fisher
>                     The plural of anecdote is not data. ~ Roger Brinner
>                     The combination of some data and an aching desire
>                     for an answer does not ensure that a reasonable
>                     answer can be extracted from a given body of data.
>                     ~ John Tukey
>
>                     2016-07-07 20:17 GMT+02:00 Joaquín Aldabe
>                     <joaquin.aldabe at gmail.com
>                     <mailto:joaquin.aldabe at gmail.com>>:
>
>                         Dear all, I´m a basic user of mixed models but
>                         there are things that I´ve
>                         found hard to understand. On of these is how
>                         the model solves the
>                         correlations among samples taken in the same
>                         site or levels of other kind
>                         of grouping variable.
>
>                         I´ve consulted Bolker (2015) in G. Fox et al
>                         Ed., Zuur et al 2009 and part
>                         of Pinheiro and Bates (2000). But still don´t
>                         understand the way that GLMM
>                         include the correlation effect in the
>                         predictions of the model.
>
>                         I really appreciate any comment and/or
>                         bibliographic reference that help me
>                         understand this.
>
>                         Thanks in advanced,
>                         Joaquín.
>
>                         --
>                         *Joaquín Aldabe*
>
>                         *Grupo Biodiversidad, Ambiente y Sociedad*
>                         Centro Universitario de la Región Este,
>                         Universidad de la República
>                         Ruta 15 (y Ruta 9), Km 28.500, Departamento de
>                         Rocha
>
>                         *Departamento de Conservación*
>                         Aves Uruguay
>                         BirdLife International
>                         Canelones 1164, Montevideo
>
>                         https://sites.google.com/site/joaquin.aldabe
>                         <https://sites.google.com/site/perfilprofesionaljoaquinaldabe>
>
>                         [[alternative HTML version deleted]]
>
>                         _______________________________________________
>                         R-sig-mixed-models at r-project.org
>                         <mailto:R-sig-mixed-models at r-project.org>
>                         mailing list
>                         https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models
>
>
>
>
>
>                 -- 
>                 *Joaquín Aldabe*
>
>                 /Grupo Biodiversidad, Ambiente y Sociedad/
>                 Centro Universitario de la Región Este, Universidad de
>                 la República
>                 Ruta 15 (y Ruta 9), Km 28.500, Departamento de Rocha
>                 /Departamento de Conservación/
>                 Aves Uruguay
>                 BirdLife International
>                 Canelones 1164, Montevideo
>
>                 https://sites.google.com/site/joaquin.aldabe
>                 <https://sites.google.com/site/perfilprofesionaljoaquinaldabe>
>
>
>
>
>             -- 
>             *Joaquín Aldabe*
>
>             /Grupo Biodiversidad, Ambiente y Sociedad/
>             Centro Universitario de la Región Este, Universidad de la
>             República
>             Ruta 15 (y Ruta 9), Km 28.500, Departamento de Rocha
>             /Departamento de Conservación/
>             Aves Uruguay
>             BirdLife International
>             Canelones 1164, Montevideo
>
>             https://sites.google.com/site/joaquin.aldabe
>             <https://sites.google.com/site/perfilprofesionaljoaquinaldabe>
>
>
>
>
>
>     -- 
>     *Joaquín Aldabe*
>
>     /Grupo Biodiversidad, Ambiente y Sociedad/
>     Centro Universitario de la Región Este, Universidad de la República
>     Ruta 15 (y Ruta 9), Km 28.500, Departamento de Rocha
>     /Departamento de Conservación/
>     Aves Uruguay
>     BirdLife International
>     Canelones 1164, Montevideo
>
>     https://sites.google.com/site/joaquin.aldabe
>     <https://sites.google.com/site/perfilprofesionaljoaquinaldabe>
>
>

-- 
Dr. Alain F. Zuur

First author of:
1. Beginner's Guide to GAMM with R (2014).
2. Beginner's Guide to GLM and GLMM with R (2013).
3. Beginner's Guide to GAM with R (2012).
4. Zero Inflated Models and GLMM with R (2012).
5. A Beginner's Guide to R (2009).
6. Mixed effects models and extensions in ecology with R (2009).
7. Analysing Ecological Data (2007).

Highland Statistics Ltd.
9 St Clair Wynd
UK - AB41 6DZ Newburgh
Tel:   0044 1358 788177
Email: highstat at highstat.com
URL:   www.highstat.com



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