Bookmark and Share

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


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

Re: st: QUAIDS with demographic translation


From   Nick Cox <njcoxstata@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: QUAIDS with demographic translation
Date   Wed, 29 Aug 2012 15:26:04 +0100

The definition is

local d1 :word count `demographics'

but no local -demographics- is defined by you in the program. So -d1-
will be zero.

This kind of error can be caught by careful line-by-line reading,
which is just how I found it.

Nick

On Wed, Aug 29, 2012 at 2:04 PM, Khalida Akhtar Choudhury
<310543@soas.ac.uk> wrote:
> Dear Nick,
>
> I've run the first few commands without the inclusion of demographic
> variables and it has worked. I think the error lies some where with
> the inclusion of local macro d1, and the inclusion of d1 in the model.
>
> Khalida
>
> On 29 August 2012 13:34, Nick Cox <njcoxstata@gmail.com> wrote:
>> This loop
>>
>> local j 1
>>
>> forvalues i = 1/$nminus1 {
>>
>> local w`i' `:word `j' of `varlist''
>>
>> local `++j'
>>
>> }
>>
>> seems unlikely to be what you intend. You don't need two macros to
>> loop over a list, and the effect of
>>
>> local `++j'
>>
>> is not that of
>>
>> local ++j
>>
>> I'd guess that you want
>>
>> forvalues i = 1/$nminus1 {
>>                   local w`i' `:word `i' of `varlist''
>> }
>>
>> In general, this is a long program and I guess that you are not an
>> experienced Stata programmer. It is best to use -set trace- and lots
>> of -display- statements to go through your program line by line.
>>
>> Nick
>>
>>
>> On Wed, Aug 29, 2012 at 12:40 PM, Khalida Akhtar Choudhury
>> <310543@soas.ac.uk> wrote:
>>> Hello,
>>>
>>> I have tried to code my own do file for QUAIDS with demographic
>>> translation. I am using the approach of Gould & Dong (2004). There are
>>> two differences between my approach and that of Banks et al. (1997).
>>> There is a scalar term representing demographic variables(ti(d)) in
>>> the welfare share equation(wi) and the expenditure m* is now dependent
>>> on demographic variables.
>>>
>>> I get error 198, "nlsurquaids returned 198 verify that nlsurquaids is
>>> a function evaluator program an error occurred when bootstrap executed
>>> estimate."
>>>
>>> set seed 1
>>>
>>> global ngoods =3D 4 // no. of goods
>>>
>>> global nbootstraps =3D 3 // no. of bootstrap replications
>>>
>>> global anot =3D 6 // alpha_0 parameter
>>>
>>>
>>>
>>>
>>> global nminus1 =3D $ngoods - 1
>>>
>>>
>>>
>>> capture program drop nlsurquaids
>>>
>>> program nlsurquaids, rclass
>>>
>>>
>>>
>>> version 10
>>>
>>>
>>>
>>> local nv =3D $ngoods + $nminus1 + 1
>>>
>>> syntax varlist(min=3D`=3D`nv'' max=3D`=3D`nv'') if, at(name)
>>>
>>>
>>>
>>> local j 1
>>>
>>> forvalues i =3D 1/$nminus1 {
>>>
>>> local w`i' `:word `j' of `varlist''
>>>
>>> local `++j'
>>>
>>> }
>>>
>>> forvalues i =3D 1/$ngoods {
>>>
>>> local lnp`i' `:word `j' of `varlist''
>>>
>>> local `++j'
>>>
>>> }
>>>
>>> local lnm `:word `j' of `varlist''
>>>
>>> local col 1
>>>
>>> local d1 :word count `demographics'
>>>
>>> // alpha
>>>
>>> forvalues i =3D 1/$nminus1 {
>>>
>>> tempname a`i'
>>>
>>> scalar `a`i'' =3D `at'[1,`col']
>>>
>>> local `++col'
>>>
>>> }
>>>
>>> tempname a${ngoods}
>>>
>>> scalar `a${ngoods}' =3D 1
>>>
>>> forvalues i =3D 1/$nminus1 {
>>>
>>> scalar `a${ngoods}' =3D `a${ngoods}' - `a`i''
>>>
>>> }
>>>
>>> // beta
>>>
>>> forvalues i =3D 1/$nminus1 {
>>>
>>> tempname b`i'
>>>
>>> scalar `b`i'' =3D `at'[1,`col']
>>>
>>> local `++col'
>>>
>>> }
>>>
>>> tempname b${ngoods}
>>>
>>> scalar `b${ngoods}' =3D 0
>>>
>>> forvalues i =3D 1/$nminus1 {
>>>
>>> scalar `b${ngoods}' =3D `b${ngoods}' - `b`i''
>>>
>>> }
>>>
>>> // gamma
>>>
>>> forvalues i =3D 1/$nminus1 {
>>>
>>> forvalues j =3D 1/$nminus1 {
>>>
>>> tempname g`i'`j'
>>>
>>> if `i' > `j' {
>>>
>>> scalar `g`i'`j'' =3D `g`j'`i''
>>>
>>> }
>>>
>>> else {
>>>
>>> scalar `g`i'`j'' =3D `at'[1, `col']
>>>
>>> local `++col'
>>>
>>> }
>>>
>>> }
>>>
>>> tempname g`i'${ngoods}
>>>
>>> scalar `g`i'${ngoods}' =3D 0
>>>
>>> forvalues j =3D 1/$nminus1 {
>>>
>>> scalar `g`i'${ngoods}' =3D  `g`i'${ngoods}' - `g`i'`j''
>>>
>>> }
>>>
>>> }
>>>
>>> forvalues j =3D 1/$nminus1 {
>>>
>>> tempname g${ngoods}`j'
>>>
>>> scalar `g${ngoods}`j'' =3D `g`j'${ngoods}'
>>>
>>> }
>>>
>>> tempname g${ngoods}${ngoods}
>>>
>>> scalar `g${ngoods}${ngoods}' =3D 0
>>>
>>> forvalues j =3D 1/${ngoods} {
>>>
>>> scalar `g${ngoods}${ngoods}' =3D `g${ngoods}${ngoods}' -  ///
>>>
>>> `g${ngoods}`j''
>>>
>>> }
>>>
>>> // lambda
>>>
>>> forvalues i =3D 1/$nminus1 {
>>>
>>> tempname l`i'
>>>
>>> scalar `l`i'' =3D `at'[1,`col']
>>>
>>> local `++col'
>>>
>>> }
>>>
>>> tempname l${ngoods}
>>>
>>> scalar `l${ngoods}' =3D 0
>>>
>>> forvalues i =3D 1/$nminus1 {
>>>
>>> scalar `l${ngoods}' =3D `l${ngoods}' - `l`i''
>>>
>>> }
>>>
>>> //omega
>>>
>>> forvalues i =3D 1/$nminus1 {
>>>
>>> tempname o`i'
>>>
>>> scalar `o`i'' =3D `at'[1,`col']
>>>
>>> local `++col'
>>>
>>> }
>>>
>>>
>>>
>>> quietly {
>>>
>>> tempvar lnpindex
>>>
>>> gen double `lnpindex' =3D $anot
>>>
>>> forvalues i =3D 1/$ngoods {
>>>
>>> replace `lnpindex' =3D `lnpindex' + `a`i''*`lnp`i''
>>>
>>> }
>>>
>>> forvalues i =3D 1/$ngoods {
>>>
>>> forvalues j =3D 1/$ngoods {
>>>
>>> replace `lnpindex' =3D `lnpindex' +  ///
>>>
>>> 0.5*`g`i'`j''*`lnp`i''*`lnp`j''
>>>
>>> }
>>>
>>> }
>>>
>>> tempvar bp
>>>
>>> gen double `bp' =3D 0
>>>
>>> forvalues i =3D 1/$ngoods {
>>>
>>> replace `bp' =3D `bp' + `lnp`i''*`b`i''
>>>
>>> }
>>>
>>> replace `bp' =3D exp(`bp')
>>>
>>> forvalues i =3D 1/$nminus1 {
>>>
>>> replace `w`i'' =3D `a`i'' + `o`i''*`demographics'
>>>
>>> forvalues j =3D 1/$ngoods {
>>>
>>> replace `w`i'' =3D `w`i'' + `g`i'`j''*`lnp`j''
>>>
>>> }
>>>
>>> replace `w`i'' =3D `w`i'' +  ///
>>>
>>> `o`i''*`demographics'+ ///
>>>
>>> `b`i''*(`lnm' - `lnpindex') + ///
>>>
>>> `l`i''/`bp'*(`lnm' - `lnpindex')^2
>>>
>>> }
>>>
>>> }
>>>
>>>
>>>
>>> end
>>>
>>>
>>>
>>> // This program fits the QUAIDS model, then it calculates the elasticities
>>>
>>> // at the sample means.
>>>
>>>
>>>
>>> capture program drop estimate
>>>
>>> program estimate, rclass
>>>
>>>
>>>
>>> // Build up nlsur command
>>>
>>> forvalues i =3D 1/$nminus1 {
>>>
>>> local vlist `vlist' w`i'
>>>
>>> local alist `alist' a`i'
>>>
>>> local blist `blist' b`i'
>>>
>>> local llist `llist' l`i'
>>>
>>> local tlist `olist' o`i'
>>>
>>> }
>>>
>>> forvalues i =3D 1/$ngoods {
>>>
>>> local vlist `vlist' lnp`i'
>>>
>>> }
>>>
>>> local vlist `vlist' lnexp
>>>
>>> forvalues i =3D 1/$nminus1 {
>>>
>>> forvalues j =3D `i'/$nminus1 {
>>>
>>> local glist `glist' g`i'`j'
>>>
>>> }
>>>
>>> }
>>>
>>> nlsur quaids @ `vlist', ifgnls nequations($nminus1) ///
>>>
>>> param(`alist' `blist' `glist' `llist'`olist')
>>> *
>>> *   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/
> *
> *   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   |   Site index