Re: st: nlsuraids need help with AIDS model error please

["Brian P. Poi" <bpoi@stata.com>]

On 07/11/2012 04:31 AM, Liu, Pei Chun wrote:
> Thank you so much Brian. It's working like a charm now!!! If I may ask another question, is the below code right if I want to calculate the AIDS with stone price index instead of translog price index?
>
> #delimit ;
> program nlsuraidstones;
>          version 11.2;
>
>    syntax varlist(min=6 max=6) if , at(name);
>          tokenize `varlist';
>          args w1 w2 lnp1 lnp2 lnp3 lnm;
>          tempname a1 a2 a3;
>          scalar `a1' = `at'[1,1];
>          scalar `a2' = `at'[1,2];
>          scalar `a3' = 1 - `a1' - `a2';
>
>          tempname b1 b2;
>          scalar `b1' = `at'[1,3];
>          scalar `b2' = `at'[1,4];
>
>          tempname g11 g12 g13;
>          tempname g21 g22 g23;
>          tempname g31 g32 g33;
>
>          scalar `g11' = `at'[1,5];
>          scalar `g12' = `at'[1,6];
>          scalar `g13' =-`g11'-`g12';
>
>
>          scalar `g21' = `g12';
>          scalar `g22' = `at'[1,7];
>          scalar `g23' = -`g21'-`g22';
>
>
>          scalar `g31' = `g13';
>          scalar `g32' = `g23';
>          scalar `g33' = -`g13'-`g23';
>
>           quietly {;
>                  tempvar lnpindex;
>                  gen double `lnpindex' = w1*`lnp1' + w2*`lnp2' +
>                                          w3*`lnp3';
>
>                  replace `w1' = `a1' + `g11'*`lnp1' + `g12'*`lnp2' +
>                          `g13'*`lnp3'+ `b1'*(`lnm' - `lnpindex');
>                  replace `w2' = `a2' + `g21'*`lnp1' + `g22'*`lnp2' +
>                          `g23'*`lnp3'+ `b2'*(`lnm' - `lnpindex');
>          };
> end;
> nlsur aidstones @ w1 w2 lnp1 lnp2 lnp3 lnm, parameters(a1 a2 b1 b2 g11 g12 g22) neq(2) ifgnls;
>
> Pei


Pei,

Your program will work fine for computing the "LA/AIDS" model with Stone's price index.

Notice that Stone's price index doesn't depend on the model parameters, so you could generate a new variable containing the price index before you even call -nlsur-.  Then inside your program you wouldn't need to keep recreating the temporary `lnpindex' variable.  Also, by using Stone's price index, the equations are in fact linear, so you could in principle use -sureg- instead of -nlsur-.  But, like I said, your program will work fine as it is written, so I wouldn't bother changing things up.

   -- Brian Poi
   -- bpoi@stata.com
*
*   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/