Statalist The Stata Listserver


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: desmat


From   "Herman van de Werfhorst" <H.G.vandeWerfhorst@uva.nl>
To   "Statalist (E-mail)" <statalist@hsphsun2.harvard.edu>
Subject   st: desmat
Date   Mon, 26 Jun 2006 10:23:45 +0200

Hello, I am using the desmat programme in stata 9 (for inclusion of different contrasts among dummy variables of one underlying ordered variable in regression models), and have difficulty with the number of parameters estimated. 

I want to use the (backward) difference contrast; which estimates regression coefficients of dummies in comparison to the previous category (rather than against one single reference category as is usually the case with dummies). 

The following .do: 

char educ[pzat] dif(b)
char p[pzat] ind(3)
desmat: regress lnwagtot @age @agesq p educ

gives five parameters for EDUC, which in total has 5 categories. This does not change if I change the first CHAR command above into

char educ[pzat] dif(b5)

If I compare the results with xi with the first category as reference category, the first four coefficients under DESMAT are to be interpreted as the difference with the previous category (i.e. the first coefficient is the difference between categories 2 and 1, the second coefficient between categories 3 and 2, etc). But I don't understand the fifth coefficient (it is not equal to the difference between 5 and 1). None of the coefficients are zero. 

Has anyone an idea? 

many thanks,
Herman van de Werfhorst






*
*   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 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index