[R-sig-eco] calculating standard error of coefficients from adonis model

[Jari Oksanen]

Rafter,

I am not Gavin, but I step in and speak for him. 

Gavin quite clearly suggested that you bootstrap adonis. The boot package is intended to generate bootstrap samples and to estimate the bootstrap confidence limits for *any* function. The boot package does not contain those functions, but you can use it to analyse adonis.

At the moment, I have no idea how to directly calculate the SEs you asked for, and even if I had, I am sure that bootstrapping would give more reliable estimates. We need to make very strong and very wrong assumptions if we try to estimate those SEs directly within adonis code, and bootstrapping is certainly better. What is sure is that the equations you gave in your message won't work.

I have never used these adonis coefficients, and I have no idea how to use them, and therefore I don't have any opinion about their use.

Cheers, Jari Oksanen
On 09/05/2014, at 03:02 AM, Rafter wrote:

> Hi Gavin,
> 
> Thanks so much for your reply. By that do you mean move away from adonis,
> and bootstrap a more conventional model that calculates parameter standard
> errors?
> 
> Adonis has been the choice for me because of its ability to handle
> multivariate, distribution-independent data in x and y. If you're suggesting
> using another model, do you have any suggestions about type? 
> 
> If you're not suggesting moving away from adonis, would perhaps slightly
> unpack your suggestion?
> 
> Thanks very much for your time and consideration.
> 
> Warmly,
> Rafter
> 
> 
> 
> --
> View this message in context: http://r-sig-ecology.471788.n2.nabble.com/calculating-standard-error-of-coefficients-from-adonis-model-tp7578878p7578882.html
> Sent from the r-sig-ecology mailing list archive at Nabble.com.
> 
> _______________________________________________
> R-sig-ecology mailing list
> R-sig-ecology at r-project.org
> https://stat.ethz.ch/mailman/listinfo/r-sig-ecology