Statalist


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

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


From   "Diego Bassani" <diego.bassani@gmail.com>
To   Statalist <statalist@hsphsun2.harvard.edu>
Subject   st: Why is there a fixed alpha for the P>|z| estimated with poisson?
Date   Fri, 12 Oct 2007 15:04:46 -0400

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/



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