[R] how to reduce stress value in isoMDS?

[stephen sefick]

no I mean maybe use a higher dimension solution-  the other thing you
may wish to check out is package vegan which has an mds function and
very good viginettes, and is my favorite package for doing ordination
analysis (metaMDS uses isoMDS).
hope this helps

On Wed, Sep 3, 2008 at 9:03 PM, 陈武 <geminiwhu at gmail.com> wrote:
> No, the two dimensions are the same. It's a dissimilarity matrix in the txt
> file.
> data[x][x]=0 data[x][y]=data[y][x] and 0<=data[x][y]<=1.
>
>
> 2008/9/3, stephen sefick <ssefick at gmail.com>:
>>
>> different dimensions?
>>
>> On Wed, Sep 3, 2008 at 8:13 AM, 陈武 <geminiwhu at gmail.com> wrote:
>> > haha...wrong code again, it's isoMDS not sammon in the 5th line.
>> > Thanks for Victor Lemes Landeiro's and Brian D. Ripley's advice.
>> >
>> >
>> > ÔÚ08-9-3£¬Prof Brian Ripley <ripley at stats.ox.ac.uk> дµÀ£º
>> >>
>> >> There is still no call to isoMDS in your code, and nothing we can
>> >> reproduce.
>> >>
>> >> It all depends on the dissimilarity matrix you have not given us: maybe
>> >> there is no good 2D representation of it.
>> >>
>> >> Looks like you need to ask a local expert about what you are doing, for
>> >> this is a statistical and not an R question.
>> >>
>> >>
>> >> On Wed, 3 Sep 2008, ³ÂÎä wrote:
>> >>
>> >> Sorry, wrong code. The right one here:
>> >>>
>> >>> library(MASS)
>> >>> cl<-read.table("e:/data.txt",header=T,sep=",")
>> >>> row.names(cl)<-colnames(cl)
>> >>> cm<-as.matrix(cl)
>> >>> loc<-sammon(cm)
>> >>> jpeg(filename="e:/plot.gif",width = 480, height = 480, units = "px",
>> >>> pointsize = 12, quality = 75, bg = "white", res = NA, restoreConsole =
>> >>> TRUE)
>> >>> plot(loc$points,type="p")
>> >>> text(loc$points,rownames(cl),cex=1,pos=1,offset=1)
>> >>> dev.off()
>> >>>
>> >>> And "e:/data.txt" contains a 40*40 dissimilarity matrix. Thanks for
>> >>> you
>> >>> advices!
>> >>>
>> >>>
>> >>> 2008/9/3, ???? <geminiwhu at gmail.com>:
>> >>>
>> >>>>
>> >>>> I apply isoMDS to my data, but the result turns out to be bad as the
>> >>>> stress
>> >>>> value stays around 31! Yeah, 31 ,not 3.1... I don't know if I ignore
>> >>>> something before recall isoMDS.
>> >>>> My code as follow:
>> >>>>
>> >>>> m <- read.table("e:/tsdata.txt",header=T,sep=",")
>> >>>> article_number <- ts(m, start = 2004,end=2008, frequency = 1
>> >>>> ,names=colnames(m))
>> >>>> jpeg(filename="e:/tsmap.gif",width = 480, height = 480, units = "px",
>> >>>> pointsize = 12, quality = 75, bg = "white", res = NA, restoreConsole
>> >>>> =
>> >>>> TRUE)
>> >>>> plot(article_number, plot.type="single",
>> >>>> lty=c(1,1,1,1,1),col=c(1,2,3,4,5),las=1)
>> >>>> max<-range(m)
>> >>>> x<-c(2004,2004,2004,2004,2004)
>> >>>>
>> >>>>
>> >>>>
>> >>>> y<-c(max[2]*0.96,max[2]*0.9199999999999999,max[2]*0.88,max[2]*0.84,max[2]*0.7999999999999999)
>> >>>> points(x,y,col=c(1,2,3,4,5),pch=15)
>> >>>> text(x,y,colnames(m),pos=4,offset=0.4)
>> >>>> dev.off()
>> >>>>
>> >>>> A 40*40 matrix in "e:/tsdata.txt". How should I do to improve the
>> >>>> effect?
>> >>>> Thank you!
>> >>>>
>> >>>>
>> >>>        [[alternative HTML version deleted]]
>> >>>
>> >>
>> >> --
>> >> Brian D. Ripley,                  ripley at stats.ox.ac.uk
>> >> Professor of Applied Statistics,  http://www.stats.ox.ac.uk/~ripley/
>> >> University of Oxford,             Tel:  +44 1865 272861 (self)
>> >> 1 South Parks Road,                     +44 1865 272866 (PA)
>> >> Oxford OX1 3TG, UK                Fax:  +44 1865 272595
>> >
>> >        [[alternative HTML version deleted]]
>> >
>> >
>> > ______________________________________________
>> > R-help at r-project.org mailing list
>> > https://stat.ethz.ch/mailman/listinfo/r-help
>> > PLEASE do read the posting guide
>> > http://www.R-project.org/posting-guide.html
>> > and provide commented, minimal, self-contained, reproducible code.
>> >
>> >
>>
>>
>>
>> --
>> Stephen Sefick
>> Research Scientist
>> Southeastern Natural Sciences Academy
>>
>> Let's not spend our time and resources thinking about things that are
>> so little or so large that all they really do for us is puff us up and
>> make us feel like gods. We are mammals, and have not exhausted the
>> annoying little problems of being mammals.
>>
>>        -K. Mullis
>
>



-- 
Stephen Sefick
Research Scientist
Southeastern Natural Sciences Academy

Let's not spend our time and resources thinking about things that are
so little or so large that all they really do for us is puff us up and
make us feel like gods. We are mammals, and have not exhausted the
annoying little problems of being mammals.

	-K. Mullis