Statalist


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

Re: st: check_eq: limited capacity of locals


From   Maarten buis <maartenbuis@yahoo.co.uk>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: check_eq: limited capacity of locals
Date   Mon, 20 Oct 2008 09:07:48 +0100 (BST)

A quick scan of the code gives a possible cause of the problem:
local p_list = r(varlist)

might better be coded as:
local p_list `r(varlist)'

But the best thing you can do is contact Rose directly (see the bottom
of the help-file), we can try to fix it, but she obviously knows the
program better than any of us and if she fixes it she can immediatly
replace the old version with a working version. 

-- Maarten

--- "Gresch,Cornelia" <gresch@mpib-berlin.mpg.de> wrote:

> Dear all,
> 
> I have a problem with the ado "check_eq" (- a package which can be
> used to check a selection of predictors for a multiple imputation
> model with (ado: ICE)):
> 
> I want to check the equations for several dependent variables by
> using 20 predictors. Since several of the predictors are categorial
> variables integrated as "dummies" into the model the different
> equations can become very large.
> 
> E.g. one of the equations looks as follows (while all *_1/*_2/*_3 ...
> variables are dummy-sets of categorial variables):
> 
> #delimit;
> check_eq, eq(
> e22106: e22107 e22104 e22105 e22103 e309m_1 e309m_2 e309m_3 e309m_4
> e309m_5 e309m_6 e309m_8 e309v_1 e309v_2 e309v_3 e309v_4 e309v_5
> e309v_6 e309v_8 e310m_1 e310m_2 e310m_3 e310m_4 e310m_5 e310m_7
> e310m_8 e339_1 e339_2 e339_4 e339_5 e339_6 e339_7 e310v_1 e310v_2
> e310v_3 e310v_4 e310v_5 e310v_7 e310v_8 e22102 tr12empr_1 tr12empr_2
> tr12empr_3 tr12empr_5 e226iv e209a_1 e209a_2 e209a_4 e209a_5 e209a_6
> e209a_7 e209a_8 tr10empr_1 tr10empr_2 tr10empr_3 tr10empr_5 e340_1
> e340_2 e340_4 e340_5 e340_6 t_sci s304_1 s304_2 s304_4 s304_5 e219a_1
> e219a_2 e219a_4 e219a_5 e219a_6 e219a_7 slsprachr_2 slsprachr_3
> slsprachr_4 slsprachr_5 slsprachr_6 slsprachr_7 slsprachr_8 e3his,
> ...-
> 
> If I use the command I get error messages like 
> 
> - e2 ambiguous abbreviation - 
> 
> I checked it with -set trace on- and the equation seems to be too
> long for one local used in the ado, at least the long list of
> predictors is cut off in the middle ending with "e2" which causes the
> error message.
> 
>  
> So my question: is there any way to either extend the capacity of
> locals and how could I do it within the ado or is there any other way
> to integrate categorical variables in a shorter way into the
> equation? It is not possible to use "stars" (like e310m_* instead of
> e310m_1 e310m_2 e310m_3...).
> 
> I hope you have any comments on this,
> Thanks!
> Cornelia
> 
> 
> **********************************************
> Cornelia Gresch (geb. Hausen)
> Max-Planck-Institut für Bildungsforschung
> Lentzeallee 94
> 14195 Berlin - Germany
> Raum: 296
> Telefon: 030-82406-675
> Email: gresch@mpib-berlin.mpg.de
> 
> 
> *
> *   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/
> 


-----------------------------------------
Maarten L. Buis
Department of Social Research Methodology
Vrije Universiteit Amsterdam
Boelelaan 1081
1081 HV Amsterdam
The Netherlands

visiting address:
Buitenveldertselaan 3 (Metropolitan), room N515

+31 20 5986715

http://home.fsw.vu.nl/m.buis/
-----------------------------------------

Send instant messages to your online friends http://uk.messenger.yahoo.com 
*
*   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 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index