# Re: st: dependent as denominator on the RHS

 From "Mizanur Rahman" To statalist@hsphsun2.harvard.edu Subject Re: st: dependent as denominator on the RHS Date Wed, 27 Jun 2007 16:23:43 +0900 (JST)

```Mizan's reply to Nguyen:
Yes the problem of simulteneity can be resolved if you have a panel data with T>=3 and a reasonable N. The ideas come from Holtz-Eakin, Newey and Rosen (1988), Arellano and Bond (1991) and Blundell and Bond (1998). What you have to do is to define a GMM-style instrument matrix for the endogenous/predetermined VARs. Please follow our David Roodman's xtabond2 module, now available as a user-written package.

type:
ssc install xtabond2, all

David also wrote an accessible discussion paper on the module.
Roodman, D. 2006. How to Do xtabond2: An Introduction to "Difference" and "System" GMM in Stata. Working Paper 103. Center for Global Development, Washington.

Nguyen will probably require sometime to get well versed with the module, if he is a first-time user of the module. But put your question on the statlistserver. Many of us not only work with xtabond2 at STATA, but also conduct GMM estimations at GAUSS and other statistical packages.

Best,
Mizan

--- "Quang Nguyen" <quangn@gmail.com> ---

>Mizan's suggestion on the use of panel data  to solve potential
>endogeneity is interesting. Can Mizan or anyone outline
>steps/procedures I can follow  with the panel data approach?
>
>Thanks and Have A Great Day!
>
>
>
>On 6/26/07, d0305@grips.ac.jp <d0305@grips.ac.jp> wrote:
>> NAM said:
>> May I have a question regarding using dependent variable as a denominator of one independent variable? I have searched your site but could not find a clear answer. To be specific, this is my model:
>> Revenue: dependent variable
>> Export ratio= Export / Revenue: independent variable
>> I treat "export ratio" as an endogenous variable and expect it to have a positive effect on "revenue".
>> Could you please tell me if I will be criticized to make my model like this?
>> I have cross-section data. Can I just use normal endogenous model to run or is there any better model?
>>
>> I would rather take Export rather than Export ratio= Export / Revenue as an independent variable and then log-tranform both the dependent var REVENUE and the EXPORT var. The point estimate would be the elasticity of REVENUE w.r.t EXPORTS. The question of endogeneity is obvious.Note that REVENUE is a linear combination of EXPORTS for a firm i. Given that you have cross-section data, I doubt if you have a reasonable set of instruments. The problem could be well resolved in the case you have had panel data.
>>
>> Mizan
>>
>>
>>
>>
>> *
>> *   For searches and help try:
>> *   http://www.stata.com/support/faqs/res/findit.html
>> *   http://www.stata.com/support/statalist/faq
>> *   http://www.ats.ucla.edu/stat/stata/
>>
>
>
>--
>"My father gave me the greatest gift anyone could give another person,
>he believed in me." - Jim Valvano
>*
>*   For searches and help try:
>*   http://www.stata.com/support/faqs/res/findit.html
>*   http://www.stata.com/support/statalist/faq
>*   http://www.ats.ucla.edu/stat/stata/

*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/
```