Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: Intruments in XTABOND


From   "Ortega, Araceli" <aorteg@essex.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: Intruments in XTABOND
Date   Mon, 18 Aug 2003 18:55:58 +0100

Dear Statalist users ;

This is my first time using the stata list, I do not know where to post
it.

My question is :

" I am estimating a model using the XTABOND command, using some
additional instrumental variables, such that the model is 

xtabond income l.Hcapital l.Pcapital l.exp, inst(tax1 tax2 tax3)

XTABOND doest not deliver the coefficients for tax1 tax2 tax3.

1. How can I recover those three coefficients?
2. How can I specify that tax1 tax2 tax3 are instruments only for l.exp.
3. I have done the same estimation using "xtivreg, fe" and "xtivreg, re"
and using the command first, you can visualise the coefficients of tax1
tax2 tax3. But I need to use Xtabond"


Thanks in advance for any help.

Araceli Ortega

-----Original Message-----
From: Nick Cox [mailto:n.j.cox@durham.ac.uk] 
Sent: 18 August 2003 18:38
To: statalist@hsphsun2.harvard.edu
Subject: Re: st: RE:mean+ci plot: how to do a ciplot with bar
representation

On Friday, Marcello Pagano posted for Eva Poen (EP) replying
to me (NJC):

EP:

>>>I would like to plot my data in a mean + confidence
>>>interval way (Stata 8).
>>>This can be nicely done with -ciplot-.
>>>My question:
>>>I would prefer to have the mean represented by a bar
>>>instead of a dot,
>>>starting from zero, just like this:
>>>
>>>    ^
>>>    |        ---
>>>    |         |
>>>    |         |
>>>    |         |
>>>    |     XXXX|XXXX
>>>    |     XXXX|XXXX
>>>    |     XXXX|XXXX
>>>    |     XXXX|XXXX
>>>    |     XXX---XXX
>>>    |     XXXXXXXXX
>>>    |     XXXXXXXXX
>>>    |     XXXXXXXXX
>>>    |     XXXXXXXXX
>>>    |     XXXXXXXXX
>>>    ---------------------->
>>>
>>>
>>>Is it possible to draw such a graph in Stata 8?

NJC:

>>It shouldn't be too difficult to write a version of
>>-ciplot- to do confidence intervals shown as these bar on bar
>>plots, but I await expressions of interest in using that
>>program before I actually write it.

EP:

>Adding a bar on bar feature to -ciplot- would be a wonderful and very
>useful extension in my opinion, and I hope others think the same.

Just to report that I hacked out a -cibplot- which Eva has tried
out privately. Certain things are wired in, because others are
against my prejudices, although anyone with the energy to read
the help file and the competence to modify the code could tweak the
program to their prejudices.

As Kit Baum is away this week, and from Thursday I'm away on vacation
for a while, it may be a time before this is up on SSC, but it now
exists.

Nick
n.j.cox@durham.ac.uk

*
*   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/



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