Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

RE: st: working with similarity matrices


From   "Clarence C. Gravlee" <[email protected]>
To   <[email protected]>
Subject   RE: st: working with similarity matrices
Date   Fri, 25 Feb 2005 16:27:19 -0500

Alan and Emmanuel,

Many thanks for your replies.  I am familiar with Ucinet and Pajek (and am a
member of INSNA) but would still prefer that my general statistical
computing platform have the ability to compute, manipulate, and save
similarity matrices.  

Almost all of Stata's competitors appear to have this ability -- SPSS, SAS,
SYSTAT, R, etc. -- and so I imagine that it must be possible in Stata, too.
If not, then it seems Matissa Hollister hit the mark in a post to Statalist
on July 27, 2004:  "Stata loses out."  

http://www.stata.com/statalist/archive/2004-07/msg00758.html

I'd like to make this work in Stata, if possible.  Can anyone please point
me in the right direction?

Thanks again for your time.

Lance Gravlee


> -----Original Message-----
> From: [email protected] 
> [mailto:[email protected]] On Behalf Of 
> Alan Neustadtl
> Sent: Friday, February 25, 2005 8:44 AM
> To: [email protected]
> Subject: Re: st: working with similarity matrices
> 
> You might have a better experience using a software package 
> designed especially for your situation.  Both Ucinet 
> (www.analytictech.com) and Pajek 
> (http://vlado.fmf.uni-lj.si/pub/networks/pajek/) can do what 
> you are asking (and more).  Older versions of Ucinet are 
> free, the latest version can be used for a limited time for 
> free, and Pajek is also free.  Both are social network 
> analysis applications.
> 
> If you decide to use Ucinet and need some help to get going, 
> let me know.
> 
> Best,
> Alan
> 
> 
> On Thu, 24 Feb 2005 22:55:32 -0500, Clarence C. Gravlee 
> <[email protected]> wrote:
> > Statalisters,
> > 
> > I have a standard case-by-variable profile matrix like this
> > 
> >   A B C D E F
> > 1
> > 2
> > 3
> > 4
> > 
> > and want to compute a case-by-case similarity matrix like this
> > 
> >   1 2 3 4
> > 1
> > 2
> > 3
> > 4
> > 
> > In this instance, I want to use Jaccard's coefficient of 
> similarity, 
> > though at other times I'll want to use another similarity 
> measure.  I 
> > need to be able to save this similarity matrix so that I 
> can work with 
> > it using multidimensional scaling, principal components 
> analysis, etc.
> > Is this sort of thing possible in Stata?
> > 
> > It appears that several people have asked related questions about 
> > working with similarity matrices in the last couple of 
> years, and no 
> > one seems to have gotten very far.  See, for example
> > 
> > http://www.stata.com/statalist/archive/2004-07/msg00605.html
> > http://www.stata.com/statalist/archive/2005-02/msg00040.html
> > 
> > Is part of the issue that Stata assumes users would want to 
> work with 
> > similarities only as part of cluster analysis?
> > 
> > As is probably apparent from my question, I don't have any 
> programming 
> > experience in Stata.  I'm willing to learn, but given my needs, I 
> > wonder if Stata is the best language to invest in learning, 
> or whether 
> > I ought to look elsewhere (like R).  I'd welcome your suggestions.
> > 
> > Thanks for your sharing your time and expertise.
> > 


*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



© Copyright 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index