[R-sig-eco] Identification of outliers using NMDS goodness of fit statistic

[scoyoc]

R-sig-ecology peeps,

I have been helping a friend run an NMDS and we started talking about the
the goodness of fit statistic (GOF) to identify outliers. It got me thinking
about this metric a bit more. I have always used exploratory analysis
procedures, like normal qqplot, histograms, box plots, ect., to identify
outlier in my data. So I have two questions regarding GOF: 1) can the NMDS
goodness of fit statistic be used to identify outliers? And 2) is there a
recommended threshold of the NMDS goodness of fit statistic that would
identify outliers?

Thanks,




-----
​
MVS
=====
Matthew Van Scoyoc
Graduate Research Assistant, Ecology
Wildland Resources Department & Ecology Center
Quinney College of Natural Resources
Utah State University
Logan, UT
=====
Think SNOW!
--
View this message in context: http://r-sig-ecology.471788.n2.nabble.com/Identification-of-outliers-using-NMDS-goodness-of-fit-statistic-tp7578632.html
Sent from the r-sig-ecology mailing list archive at Nabble.com.