Re: st: Cluster analysis variables - bug?

[Steven Joel Hirsch Samuels <sjhsamuels@earthlink.net>]

--

The name option names  group variable.  In Stata 9.2, only one group  
variable is created at a time.  Did you run a previous cluster  
analysis on the same data, possibly different variables, with three  
groups and "name(cluster)"?   If so,  that would account for your  
finding.

--Steve

Example:

. sysuse auto
(1978 Automobile Data)

. cluster kmeans mpg weight displacement, k(3) name(cluster01)

. cluster list cluster01
cluster01  (type: partition,  method: kmeans,  dissimilarity: L2)
      vars: cluster01 (group variable)
     other: k: 3
            start: krandom
            range: 0 .
            cmd: cluster kmeans mpg weight displacement, k(3) name 
(cluster01)
            varlist: mpg weight displacement


. tab cluster01

  cluster01 |      Freq.     Percent        Cum.
------------+-----------------------------------
          1 |         15       20.27       20.27
          2 |         39       52.70       72.97
          3 |         20       27.03      100.00
------------+-----------------------------------
      Total |         74      100.00

. ds
make          mpg           headroom      weight        turn           
gear_ratio    cluster01
price         rep78         trunk         length        displacement   
foreign





On Aug 8, 2007, at 12:56 PM, Ricardo Ovaldia wrote:


> Thank you Nick. But I am not sure that helps. Both
> variables are discrete taking the values 1, 2 and 3.
> (I specific 3 clusters). Every value of -C1- appears
> for every value of -cluster-. I am at lost.
>
>
>            |                c1
>    cluster |         1          2          3 |
> Total
> -----------+---------------------------------+----------
>          1 |        71         90         17 |
> 178
>          2 |         4        200        596 |
> 800
>          3 |       123        164          5 |
> 292
> -----------+---------------------------------+----------
>      Total |       198        454        618 |
> 1,270
>
>
> > --- Nick Cox <n.j.cox@durham.ac.uk> wrote:
> >
> > > A scatter plot of the two variables may
> > > throw light on this.
> > >
> > > Nick
> > > n.j.cox@durham.ac.uk
> > >
> > > Ricardo Ovaldia
> > >
> > > > This may be a very simple question, but I can't
> > > find
> > > > the answer. Doing a cluster analysis I typed:
> > > >
> > > > . cluster kmeans  q4_*, k(3) name(c1)
> > s(kr(67492))
> > > > I get two new variables -c1- and -cluster-.
> > These
> > > are
> > > > different. Could someone please tell me what is
> > > the
> > > > difference between these two variables? Which
> > one
> > > is
> > > > the cluster indicator.
> > > *
> > > *   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/
> > Ricardo Ovaldia, MS
> > Statistician
> > Oklahoma City, OK
> ______________________________________________________________________ 
> ______________
>
> > Need a vacation? Get great deals
> > to amazing places on Yahoo! Travel.
> > http://travel.yahoo.com/
> > *
> > *   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/
>
> Ricardo Ovaldia, MS
> Statistician
> Oklahoma City, OK
>
>
>        
> ______________________________________________________________________ 
> ______________
> Fussy? Opinionated? Impossible to please? Perfect.  Join Yahoo!'s  
> user panel and lay it on us. http://surveylink.yahoo.com/gmrs/ 
> yahoo_panel_invite.asp?a=7
>
> *
> *   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/
Steven Joel Hirsch Samuels

sjhsamuels@earthlink.net
18 Cantine's Island
Saugerties, NY 12477
Phone: 845-246-0774
EFax: 208-498-7441




*
*   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/