RE: st: 2-stages cluster analysis

[Matt Loke <sciacallojo@hotmail.com>]

I know that
it was a pseudo-syntax. Nevertheless, in order to fill in the gaps in it, I need to figure out how does the syntax you propose work. Indeed, I do not understand very well the loop... how can I narrow the cluster analysis (in the second stage) to each group created at the first stage? 


Thanks,
Matt

----------------------------------------
> Date: Mon, 20 Sep 2010 23:23:41 -0500
> Subject: Re: st: 2-stages cluster analysis
> From: skolenik@gmail.com
> To: statalist@hsphsun2.harvard.edu
>
> On Mon, Sep 20, 2010 at 11:10 PM, Matt Loke  wrote:
>> Thanks for your help. Nevertheless, I need you to explain me the command lines you posted. Indeed, the command cluster creates three variables (stage1cl_id, stage1cl_ord and stage1cl_hgt). I thought that the one you were referring to was stage1cl_hgt. Anyway, the loop gives me a syntax error.
>
> Sure... it was a pseudo-syntax, you need to fill in gaps in it.
> Anyways, you are doing the hierarchical analysis, and the variables
> you see provide the description of the tree. You need to cut your tree
> to obtain your clusters:
>
> cluster generate stage1clusters = cut(), name( stage1 )
>
> and then you need to do your cluster analysis within the groups
> identified by this new variable.
>
> --
> Stas Kolenikov, also found at http://stas.kolenikov.name
> Small print: I use this email account for mailing lists only.
>
> *
> * For searches and help try:
> * http://www.stata.com/help.cgi?search
> * http://www.stata.com/support/statalist/faq
> * http://www.ats.ucla.edu/stat/stata/
 		 	   		  
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/