[R-sig-ME] spatial auto-correlation or more complicated pseudo-replication?

[Thomas Merkling]

Thanks Thierry, that's reassuring.

One last question: how do the covariates take care of the potential 
spatial autocorrelation in the response variable?

Kind regards,
Thomas

On 23/04/2020 11:44, Thierry Onkelinx wrote:
> Dear Thomas,
>
> I'd say that there is nothing that suggests spatial autocorrelation in 
> the residuals nor in the random effects. There might be spatial 
> autocorrelation in the response variable, but that is handled by the 
> covariates in the model. The pair random effect takes care of the 
> pseudo-replication.
>
> IMHO you can't state that the treatment of the focal pair influences 
> the surrounding pairs. If that was the case, I'd expect to see spatial 
> autocorrelation.
>
> Best regards,
>
> ir. Thierry Onkelinx
> Statisticus / Statistician
>
> Vlaamse Overheid / Government of Flanders
> INSTITUUT VOOR NATUUR- EN BOSONDERZOEK / RESEARCH INSTITUTE FOR NATURE 
> AND FOREST
> Team Biometrie & Kwaliteitszorg / Team Biometrics & Quality Assurance
> thierry.onkelinx using inbo.be <mailto:thierry.onkelinx using inbo.be>
> Havenlaan 88 bus 73, 1000 Brussel
> www.inbo.be <http://www.inbo.be>
>
> ///////////////////////////////////////////////////////////////////////////////////////////
> 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
> ///////////////////////////////////////////////////////////////////////////////////////////
>
> <https://www.inbo.be>
>
>
> Op wo 22 apr. 2020 om 23:24 schreef Thomas Merkling 
> <thomasmerkling00 using gmail.com <mailto:thomasmerkling00 using gmail.com>>:
>
>     Dear Thierry,
>
>     Thanks a lot for your precious help.
>
>     I changed the width argument of variogram() and obtained similar
>     patterns (for egg
>     <https://drive.google.com/open?id=1Ktinx0gm4sRNS6_r5VEAg_pdND51eHQd>
>     and for laydate
>     <https://drive.google.com/open?id=1YoMIK1jax1dhBe94yswbHvWF_CyKDNL5>).
>     Doing the same thing with the random intercepts gave similar
>     patterns too (for egg
>     <https://drive.google.com/open?id=1fGlXObzhRj2g9J9ZnYIO8V6q6d-PmgS8>
>     and for laydate
>     <https://drive.google.com/open?id=1U4Q9h7foZM6q54_yteFAFHo6SVHby9mJ>).
>
>     Can I then conclude that there is no need to correct for spatial
>     auto-correlation?
>
>     Given that treatment of a focal pair influences the value of
>     another predictor for surrounding pairs, is there any other random
>     effect that I should add? Or is this test for spatial
>     auto-correlation enough?
>
>     Kind regards,
>     Thomas
>
>     On 22/04/2020 19:45, Thierry Onkelinx wrote:
>>     Dear Thomas,
>>
>>     Have a look at the data.frame in the variogram() output. Given
>>     your variogram I expect a high number of pairs (np variable) at
>>     short range and a low (< 100) at large ranges. Note the width and
>>     cutoff arguments of variogram().  The defaults are 1/3 of the
>>     diagonal of the bounding box for cutoff and cutoff/15 for width.
>>     These are likely suboptimal for your data. I'd set width to
>>     slightly larger than the distance between two adjacent nests.
>>     Increase the width if the variogram is unstable.
>>     If you still get a similar picture as the ones you send, then
>>     there then residuals are iid and thus you don't need to correct
>>     for spatial autocorrelation.
>>
>>     Given the strong correlation between pair and location, the pair
>>     random effect will take up some of the spatial autocorrelation.
>>     You could make a variogram of the random intercepts. There should
>>     be a pure nugget effect too.
>>
>>     Best regards,
>>
>>     ir. Thierry Onkelinx
>>     Statisticus / Statistician
>>
>>     Vlaamse Overheid / Government of Flanders
>>     INSTITUUT VOOR NATUUR- EN BOSONDERZOEK / RESEARCH INSTITUTE FOR
>>     NATURE AND FOREST
>>     Team Biometrie & Kwaliteitszorg / Team Biometrics & Quality
>>     Assurance
>>     thierry.onkelinx using inbo.be <mailto:thierry.onkelinx using inbo.be>
>>     Havenlaan 88 bus 73, 1000 Brussel
>>     www.inbo.be <http://www.inbo.be>
>>
>>     ///////////////////////////////////////////////////////////////////////////////////////////
>>     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
>>     ///////////////////////////////////////////////////////////////////////////////////////////
>>
>>     <https://www.inbo.be>
>>
>>>>             .....
>>>

	[[alternative HTML version deleted]]