Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: Creating an index


From   Maarten Buis <maartenlbuis@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Creating an index
Date   Wed, 16 May 2012 09:27:36 +0200

On Tue, May 15, 2012 at 10:53 PM, Ozgur Ozdemir wrote:
> it seems, I need to use
> sheaf coefficient like technique.  "Combining information from multiple variables using models for casual indicators" explains how several variables can be grouped and used as an independent variable but not sure how I can use the new variable as a dependent variable in a test ?
> in your example, if I want to use the family socio economic status as dependent variable in a regression, first I need to create the index and calculate the actual socio economic values.

You can use the -post- option in -sheafcoef- and than use -predict- to
predict the latent variable.

*---------------------- begin example -----------------------------
sysuse nlsw88, clear
tab occupation, gen(occ)
gen ttl_exp2 = ttl_exp*ttl_exp
		
glm wage occ2-occ13 married never_married grade ttl_exp*, link(log)

// with the post option it turns -sheafcoef- into a regular
// estimation command,
sheafcoef, latent(occ: occ2 occ3 occ4 occ5 occ6 occ7 occ8 occ9 ///
                       occ10 occ11 occ12 occ13) post

// which means we can use -predict- to create the index
predict occ_status, xb equation(#2)

// and later use it in a regression as the dependent variable
reg occ_status grade i.race
*----------------------- end example ------------------------------
(For more on examples I sent to the Statalist see:
http://www.maartenbuis.nl/example_faq )

You might also consider going for a complete SEM model.

>there might be also a direction problem such as some variables might increase the index while they are decreasing but others might increase the index while they are increasing ?

That is not a problem, these two types of variables will automatically
get opposite signs on their weights.

-- Maarten

--------------------------
Maarten L. Buis
Institut fuer Soziologie
Universitaet Tuebingen
Wilhelmstrasse 36
72074 Tuebingen
Germany


http://www.maartenbuis.nl
--------------------------

*
*   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/


© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index