# st: RE: xtabond2 and parameter restrictions

 From "Gielis Wim" <[email protected]> To <[email protected]> Subject st: RE: xtabond2 and parameter restrictions Date Thu, 7 Jul 2005 10:47:56 +0200

```Hi Statalisters,

I think you can do this by stating:

define 1 xy = z
define 2 xp = q

(I abbreviate the variable names into p, q, x, y, z for convenience)

After that, you type the command for xtabond2, with

const(1 2)

BUT (I face a similar problem)

1) what to do if the restrictions you want to impose involve parameters
outside the relation (e.g. a coefficient represents the ratio of a
coefficient in the relation and the price elasticity of demand)???

2) going further, how can you obtain point estimates, standard errors,
etc. for all these variables (so including price elasticity of
demand)???

Regards,

Wim

-----Original Message-----
From: [email protected]
[mailto:[email protected]] On Behalf Of julian
estevez
Sent: donderdag 7 juli 2005 2:23
To: [email protected]
Subject: st: xtabond2 and parameter restrictions

Dear Statalisters:

I am estimating a Dynamic Panel Data model using xtabond2

It is of the form:

Dependent variable(i,t)  = constant +b1 X1(i,t) + b2 X1(i,t-1)+ b3
X2(i,t) +
b4 X2(i,t-1) + b5 Dependent variable(i,t-1) + disturbance

I wonder how I can estimate the model but including the following
parameter
restrictions:

b5 b1 = b2
b5 b3 = b4

Ignoring these restrictions mean that I am loosing information because I
am overparametrizing the model. I would greatly appreciate if somebody
could tell me how to combine parameter restrictions with xtabond2, or
some other solution to my problem

Regards

Julian

_________________________________________________________________
FREE!
http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/

*
*   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/
```