Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: RE: Dynamic Random Effects Probit Model in Stata 8.2


From   "Nick Cox" <n.j.cox@durham.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: RE: Dynamic Random Effects Probit Model in Stata 8.2
Date   Tue, 3 May 2005 13:38:19 +0100

I cannot advise on this kind of model or 
how to implement it, but in the first
instance the problem may be one of 
Stata syntax. 

As the error message shows, -xtprobit- is 
expecting a matrix here. Your -orgmb0- (or
your -Orgmb0-) is a variable, not a matrix. 

In general, it helps enormously to see 
the _exact_ command syntax used (including 
exact spelling). 

Nick 
n.j.cox@durham.ac.uk 

gmc107@york.ac.uk
 
>  I am applying to the British Household Panel Survey data set 
> a variant of
> Vella and Verbeek's dynamic model of unionism and wage rate 
> determination
> employed in "Whose Wages Do Unions Raise? A Dynamic Model of 
> Unionism and
> Wage Determination for Young Men", Journal of Applied 
> Econometrics, 1998.
> To deal with the initial conditions problem I am following J 
> Wooldridge's
> solution given in "Simple Solutions to the Initial Conditions 
> Problem in
> Dynamic, Nonlinear Panel Data Models with Unobserved 
> Heterogeneity", Journal
> of Applied Econometrics, 20, pp.20-39 (2005).
> I have attempted to implement this approach (for the reduced 
> form union
> membership determination equation);in the model provided 
> below new_orgmb0
> (eg unioni0 in Wooldridge's example) in other words the 
> initial value of
> trade union membership is not included.
> This is because I have not really been able to figure out how 
> to specify the
> entry into the Initial values box in Stata 8.2 under the maximisation
> options heading. The way I attempted to deal with it 
> practically was to
> multiply the union membership (indicator variable) by a time 
> dummy eg say
> Orgmb0=new_orgmb*T97 (where T97 is a dummy variable taking a 
> value of one if
> year=1997 eg the initial annum in my data set) and then 
> insert Orgmb0 in the
> box for  Initial values(max options) so that the simple 
> xtprobit command
> below would end with re from(orgmb0).
> This apparently does not work since I get:
> from(): matrix orgmb0 not found
> 
> r(111)
> 
> ==============================================================
> =========
> 
> T=97,98,99,00,01,02; TO=97,T1=98...,T5=02
> 
> 
> ==============================================================
> ==========
>  tsset pid year, yearly
>        panel variable:  pid, 10016848 to 77605489
>         time variable:  year, . to 0102, but with gaps
> xtprobit new_orgmb new_orgmblag logexper married married1 
> married2 married3
> married4 married5 female black  asian otherethnic 
> lonsoutheast scotland
> wales northwest northeast midlands uniqual vocqual  alevels  hlstatdum
> hlstatdum1 hlstatdum2 hlstatdum3 hlstatdum4 hlstatdum5 ftemp 
> ftemp1 ftemp2
> ftemp3 ftemp4 ftemp5  agforfish exctmanu meteng manuind 
> constr distr tancom
> banfin othesevsic profocc assproftech clersec craft perpro 
> plantmach otheocc
> T98 T99 T00 T01 T02, re
> ==============================================================
> ===========
> 
> Would be most grateful if you could please help me with this.

*
*   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