Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: Re: Why is there a fixed alpha for the P>|z| estimated with poisson?


From   "josemaria" <jmpsouza@usp.br>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: Re: Why is there a fixed alpha for the P>|z| estimated with poisson?
Date   Fri, 12 Oct 2007 22:53:31 -0300

the p-value is the Wald test ( in general (x-E(e))/se(x)) or some transformation) and does not depend of the level proposed for the CI.
José Maria
José Maria Pacheco de Souza, Professor Titular (aposentado)
Departamento de Epidemiologia
Faculdade de Saúde Pública/Universidade de São Paulo
Av. Dr. Arnaldo, 715 cep 01246-904
São Paulo Brasil
fones (11)3061-7747 (11)3061-7724 (11)3714-2403 (11)3768-8612
fax (11)3082-2920 (11)3714-2403
www.fsp.usp.br/~jmpsouza
jmpsouza@usp.br

----- Original Message ----- From: "Diego Bassani" <diego.bassani@gmail.com>
To: "Statalist" <statalist@hsphsun2.harvard.edu>
Sent: Friday, October 12, 2007 4:04 PM
Subject: st: Why is there a fixed alpha for the P>|z| estimated with poisson?



Dear Statalist members,

I have realized that the p-value for the null hypothesis (regression
coefficient = zero given that the rest of the predictors are in the
model) does not change when level (CI) is altered.
Why is the alpha for the p-value calculation in the z distribution
fixed, and is this true for other regression models as well?

Thanks

Diego

______________________
see below three examples with level set at 99, 90 and 95.

xi3: poisson dep i.ind, vce(cluster _set) irr level(99)

Iteration 0: log pseudolikelihood = -21326.6
Iteration 1: log pseudolikelihood = -21326.6

Poisson regression Number of obs = 39500
Wald chi2(1) = 3.03
Prob > chi2 = 0.0819
Log pseudolikelihood = -21326.6 Pseudo R2 = 0.0001

(Std. Err. adjusted for 8351 clusters in _set)
------------------------------------------------------------------------------
| Robust
dep| IRR Std. Err. z P>|z| [99% Conf. Interval]
-------------+----------------------------------------------------------------
_Iind | 1.034227 .0200087 1.74 0.082 .9839511 1.087072
------------------------------------------------------------------------------

. xi3: poisson dep i.ind, vce(cluster _set) irr level(95)

Iteration 0: log pseudolikelihood = -21326.6
Iteration 1: log pseudolikelihood = -21326.6

Poisson regression Number of obs = 39500
Wald chi2(1) = 3.03
Prob > chi2 = 0.0819
Log pseudolikelihood = -21326.6 Pseudo R2 = 0.0001

(Std. Err. adjusted for 8351 clusters in _set)
------------------------------------------------------------------------------
| Robust
dep | IRR Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
_Iind | 1.034227 .0200087 1.74 0.082 .9957449 1.074196
------------------------------------------------------------------------------

. xi3: poisson dep i.ind vce(cluster _set) irr level(90)

Iteration 0: log pseudolikelihood = -21326.6
Iteration 1: log pseudolikelihood = -21326.6

Poisson regression Number of obs = 39500
Wald chi2(1) = 3.03
Prob > chi2 = 0.0819
Log pseudolikelihood = -21326.6 Pseudo R2 = 0.0001

(Std. Err. adjusted for 8351 clusters in _set)
------------------------------------------------------------------------------
| Robust
dep | IRR Std. Err. z P>|z| [90% Conf. Interval]
-------------+----------------------------------------------------------------
_Iind | 1.034227 .0200087 1.74 0.082 1.001834 1.067668
------------------------------------------------------------------------------
*
* 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/



--
No virus found in this incoming message.
Checked by AVG Free Edition.
Version: 7.5.488 / Virus Database: 269.14.9/1067 - Release Date: 10/12/aaaa 18:02


*
*   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