Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, statalist.org is already up and running.

# st: QUAIDS nlsur with censoring

 From nathalie me-nsope To statalist@hsphsun2.harvard.edu Subject st: QUAIDS nlsur with censoring Date Wed, 6 Mar 2013 10:17:32 -0500

```Dear STATA Helpers,

I  am   estimating the QUAIDS model using nlsur quaids (Poi, 2008)
for a system of four goods and have non-consumption as an
issue to deal with (meaning I can't use the new QUAIDS command).
Idid not make any adjustments to the right hand side(budget shares). I
have I household variable and two sets of dummy variables For
illustrative purposes, the first line of my nlsur quaids program files
and the share equations  are provided below. After  the estimation, I
obtained some very weird parameters (some are quite
large) and so are the elasticities.

capture program drop nlsurquaidsM
program nlsurquaidsM
version 12.1

syntax varlist(min=24 max=24) if, at(name)

tokenize `varlist'
args w1 w2 w3 lnp1 lnp2 lnp3 lnp4 lnm lnAE Pr1 d1 d2 d3 d4 d5 d6 d7
d8  pdf1 pdf2 pdf3 cdf1 cdf2 cdf3

// The expenditure shares for 3 of the 4
// just one share equation  here
replace `w1' = (`a1' + `g11'*`lnp1' + `g12'*`lnp2' + ///
`g13'*`lnp3' + `g14'*`lnp4' + ///
`b1'*(`lnm' - `lnpindex') + ///
`l1'/`bofp'*(`lnm' - `lnpindex')^2 + ///
`r11'*`lnAE' +`r12'*`Pr1' +`r13'*`d1'+`r14'*`d2' +`r15'*`d3'
+`r16'*`d4'+ ///
`r17'*`d5'+`r18'*`d6'+`r19'*`d7'+`r110'*`d8')*`cdf1'+ ///
`d1'*`pdf1'

I don't know what I am doing wrong.  The discussion on Statalist on
multiplying the left and right hand side variables  by the CDF and
adding the pdf caught my attention. However, I did not seem to grasp
what was concluded  to be the correct approach. Please can you clarify
me on this and if possible share any  codes for estimation and
computation of elasticities that might help
understanding.

Many thanks