Re: [R] plot hclust - canberra dist + median linkage
Dan == Dan Bolser [EMAIL PROTECTED] on Mon, 11 Oct 2004 16:21:53 +0100 (BST) writes: Dan Gives strange results. Dan I get 'weird' dendrograms with canberra / binary distance metric and Dan median / centroid cluster methods. it doesn't depend on the metric: Both 'median' and 'centroid' methods are known to *not* guarantee ``monotone distance measures'', or equivalently to possibly lead to dendrograms with so called ``inversions''. We should add this to help page for hclust(). Probably for this reason, agnes() from the cluster package doesn't have these two methods [explicitly] -- though it now allows general parameter Lance-William formula methods which can also lead to inversions. Dan Is this just my data? evidently not. Though the problem does not appear for all data sets... Regards, Martin Maechler __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] plot hclust - canberra dist + median linkage
On Tue, 12 Oct 2004, Martin Maechler wrote: Dan == Dan Bolser [EMAIL PROTECTED] on Mon, 11 Oct 2004 16:21:53 +0100 (BST) writes: Dan Gives strange results. Dan I get 'weird' dendrograms with canberra / binary distance metric and Dan median / centroid cluster methods. it doesn't depend on the metric: Both 'median' and 'centroid' methods are known to *not* guarantee ``monotone distance measures'', or equivalently to possibly lead to dendrograms with so called ``inversions''. We should add this to help page for hclust(). Cool, I could do with more to read about these methods. It might be worth noteing that no bootstrap exists for the method (implemented in R that is). How can I quickly run a test on a sub-set of x? (i.e. quick bootstrap by hand)? Cheers, Dan. Probably for this reason, agnes() from the cluster package doesn't have these two methods [explicitly] -- though it now allows general parameter Lance-William formula methods which can also lead to inversions. Dan Is this just my data? evidently not. Though the problem does not appear for all data sets... Regards, Martin Maechler __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] plot hclust - canberra dist + median linkage
Gives strange results. I get 'weird' dendrograms with canberra / binary distance metric and median / centroid cluster methods. Is this just my data? Dan __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html