Statalist


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

Antwort: st: Censored demand estimation


From   Richard Ochmann <rochmann@diw.de>
To   statalist@hsphsun2.harvard.edu
Subject   Antwort: st: Censored demand estimation
Date   Wed, 3 Feb 2010 20:46:01 +0100

have you tried that with -sureg-?
in my experience, you can estimate a system of share equations, where the 
shares add-up to one over all G equations, with -sureg- for all G 
equations in case there is variation in the covariates over the equations.

but maybe it will be easier for you to trick the -nlsur quaids- routine 
(-net install st0029_i-) to take up all G equations. that may depend on 
your demand system, especially how similar it is to the quaids. i dont 
have any experience with -nlsur quaids- though.

best, rich

owner-statalist@hsphsun2.harvard.edu schrieb am 03.02.2010 17:02:15:

> Dear Stata Users
>  
> Household budget survey data for demand system estimation often 
> creates a major problem that is due to recording zero expenditure 
> for food aggregates that are not consumed during the survey period. 
> This cause censored dependent variables and leads to biased results 
> when not accounted for. Heien and Wessells (1990) introduced a 
> computationally simple, two-step estimation procedure based on 
> Heckman?s (1979) work. However, Shonkwiler and Yen (1999) 
> demonstrated the inconsistent two-step estimation procedure. They 
> correct for inconsistency by defining the second-stage regression...
> defined PDF that can be re-incorporated in the second stage. In 
> doing so the right hand side system of demand equations may not 
> necessary added up to 1 (unlike the usual system of demand 
> equations) and thus the equation must be estimated for full n-
> vectors (in contrast to n-1 vectors).  The usual demand system 
> routine written in STATA, take in to account for n-1 vectors
>  (including Poi?s, 2008: demand system update). I couldn?t find 
> demand system for full vector... Please share with me if you have 
> STATA routine for full n-vectors or give me some kinds of tip to go 
> head. Any help is appreciated and acknowledged
>  
> Cheers
> Nigussie 
> 
> 
> 
> 
> *
> *   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/



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