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]

st: nlsur function evaluator program QUAIDS with demographics


From   RD <r_distante@yahoo.it>
To   "statalist@hsphsun2.harvard.edu" <statalist@hsphsun2.harvard.edu>
Subject   st: nlsur function evaluator program QUAIDS with demographics
Date   Fri, 8 Mar 2013 15:39:24 +0000 (GMT)

Dear Statalist-ers,

I am trying to write a function evaluator program using -nlsur- in Stata 12 to estimate a QUAIDS model (Banks et al., Restat 1997) with household demographics. So far, I had no success, as Stata returns an error message:

'nlsurtry returned 199
    - di as error "verify that nlsur`eqn' is a function evaluator program"
    = di as error "verify that nlsurtry is a function evaluator program"
verify that nlsurtry is a function evaluator program
    - exit _rc'

I do not understand what I am doing wrong.

Also, copying the nlsurquaids program (st0029_1) by Brian Poi (2008) - which works properly - in a .ado file with a different name, placed in the same directory or in other directories "sysdir set", Stata returns the same exact error. I am puzzled.

Another silly question: is it possible to write a function evaluator program including, e.g., 15 "potential" demographic variables, but practically feeding it with, say, 7? I was wondering it because otherwise I should write different programs for different sets of demographics. 

Last, but not least, when I use -nlcom- to compute elasticities, should I type variable mean values (for the full sample or for a subsample) to be multiplied to the coefficients or it is allowed to type variable names? 

Here follows the code, slightly modified to include only one demographic variable (z1), which I plan to enlarge with more demographics.

Thanks in advance for your help.


// Demographics z_k enter as taste-shifters into the share equations and, 
// to mantain integrability, are part of the a_i terms in the price index:
// ln[a(p)] = a_0 + sum_i[a_i + sum_k(a_ik*z_k)]*ln(p_i) +
// 1/2*sum_isum_j[g_ij*ln(p_i)*ln(p_j)]
// the model in share form looks like:
// w_i = a_i + sum_k(a_ik*z_k) + sum_j[g_ij*ln(p_j)] + b_i*ln[m/a(p)] +
// [l_i/b(p)]*ln[m/a(p)]^2 
// where w_i=share of good i, p_i=price of good i, etc.
// b(p)=prod_i(p_i^b_i)

program nlsurtry

version 10
syntax varlist(min=9 max=9) if, at(name)
tokenize `varlist'
args w1 w2 w3 lnp1 lnp2 lnp3 lnp4 lnm z1 
tempname a1 a2 a3 a4
scalar `a1' = `at'[1,1]
scalar `a2' = `at'[1,2]
scalar `a3' = `at'[1,3]
scalar `a4' = 1 - `a1' - `a2' - `a3'
tempname az11 az12 az13 az14
scalar `az11' = `at'[1,4]
scalar `az12' = `at'[1,5]
scalar `az13' = `at'[1,6]
scalar `az14' = -`az11' - `az12' - `az13'
tempname b1 b2 b3 b4
scalar `b1' = `at'[1,7]
scalar `b2' = `at'[1,8]
scalar `b3' = `at'[1,9]
scalar `b4' = -`b1' - `b2' - `b3'

tempname g11 g12 g13 g14
tempname g21 g22 g23 g24
tempname g31 g32 g33 g34
tempname g41 g42 g43 g44
scalar `g11' = `at'[1,10]
scalar `g12' = `at'[1,11]
scalar `g13' = `at'[1,12]
scalar `g14' = -`g11' - `g12' - `g13'

scalar `g21' = `g12'
scalar `g22' = `at'[1,13]
scalar `g23' = `at'[1,14]
scalar `g24' = -`g21' - `g22' - `g23'

scalar `g31' = `g13'
scalar `g32' = `g23'
scalar `g33' = `at'[1,15]
scalar `g34' = -`g31' - `g32' - `g33'

scalar `g41' = `g14'
scalar `g42' = `g24'
scalar `g43' = `g34'
scalar `g44' = -`g41' - `g42' - `g43'
tempname l1 l2 l3 l4

scalar `l1' = `at'[1,16]
scalar `l2' = `at'[1,17]
scalar `l3' = `at'[1,18]
scalar `l4' = -`l1' - `l2' - `l3'

quietly {
tempvar lnpindex
gen double `lnpindex' = 5 + `a1'*`lnp1' + `a2'*`lnp2' + `a3'*`lnp3' + `a4'*`lnp4' + `az11'*`z1'*`lnp1' + `az12'*`z1'*`lnp2' + `az13'*`z1'*`lnp3' + `az14'*`z1'*`lnp4' 

forvalues i = 1/4 {
forvalues j = 1/4 {
replace `lnpindex' = `lnpindex' + 0.5*`g`i'`j''*`lnp`i''*`lnp`j''
          }
  }
tempvar bofp
gen double `bofp' = 0
forvalues i = 1/4 {
replace `bofp' = `bofp' + `lnp`i''*`b`i''
 }
replace `bofp' = exp(`bofp')
replace `w1' = `a1' + `az11'*`z1' + `g11'*`lnp1' + `g12'*`lnp2' +  `g13'*`lnp3' + `g14'*`lnp4' + `b1'*(`lnm' - `lnpindex')  +  `l1'/`bofp'*(`lnm' - `lnpindex')^2
replace `w2' = `a2' + `az12'*`z1' + `g21'*`lnp1' + `g22'*`lnp2' +  `g23'*`lnp3' + `g24'*`lnp4' + `b2'*(`lnm' - `lnpindex')  +  `l2'/`bofp'*(`lnm' - `lnpindex')^2
replace `w3' = `a3' + `az13'*`z1' + `g31'*`lnp1' + `g32'*`lnp2' + `g33'*`lnp3'  + `g34'*`lnp4' + `b3'*(`lnm' - `lnpindex')  +  `l3'/`bofp'*(`lnm' - `lnpindex')^2

}
end          

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/


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