[R-sig-eco] vegan Mantel test

[Jari Oksanen]

On 10/06/11 00:12 AM, "Scott Chamberlain" <scttchamberlain4 at gmail.com>
wrote:

> How is it possible that the upper confidence limit of mantel's r is below that
> of the estimate r value itself?
> 
> For example, using the example in the vegan documentation:
> require(vegan)
> data(varespec) data(varechem) veg.dist <- vegdist(varespec) # Bray-Curtis
> env.dist <- vegdist(scale(varechem), "euclid") mantel(veg.dist, env.dist)
> Mantel statistic based on Pearson's product-moment correlation Call:
> mantel(xdis = veg.dist, ydis = env.dist) Mantel statistic r: 0.3047
> Significance: 0.001 Empirical upper confidence limits of r: 90% 95% 97.5% 99%
> 0.110 0.145 0.173 0.196 Based on 999 permutations
> 
> 
> The 95% confidence limit is below that of the 0.3047 estimate. I must be
> reading this wrong. Is the 95% limit supposed to be the amount +/- of the
> estimate of r? 
> 
Perhaps your observed value is "significant" in which case it is outside 95%
of the permuted values?

Cheers, Jari Oksanen