Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: what is the difference between newey and nwest?


From   gjhxmu@sina.com
To   statalist<statalist@hsphsun2.harvard.edu>
Subject   st: what is the difference between newey and nwest?
Date   Fri, 31 Jul 2009 18:16:42 +0800

-newey- and -nwest- are both used to estimate Newey-West standard errors in the stata.
Why the results of them are different? 
And the results for -nwest- is different each time I run.

The results are as follows,

. nwest reg y x, lag(1) force

Regression with Newey-West standard errors          Number of obs  =      5000
maximum lag : 1                                     F(  1,  4998)  =   1320.16
                                                    Prob > F       =    0.0000

------------------------------------------------------------------------------
             |             Newey-West
           y |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
           x |   1.034833   .0284812    36.33   0.000     .9789979    1.090669
       _cons |   .0296797   .0282062     1.05   0.293    -.0256169    .0849763
------------------------------------------------------------------------------

. nwest reg y x, lag(1) force

Regression with Newey-West standard errors          Number of obs  =      5000
maximum lag : 1                                     F(  1,  4998)  =   1317.95
                                                    Prob > F       =    0.0000

------------------------------------------------------------------------------
             |             Newey-West
           y |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
           x |   1.034833    .028505    36.30   0.000     .9789512    1.090716
       _cons |   .0296797   .0282159     1.05   0.293    -.0256359    .0849953
------------------------------------------------------------------------------

. newey y x,lag(1) force

Regression with Newey-West standard errors          Number of obs  =      5000
maximum lag: 1                                      F(  1,  4998)  =   1094.36
                                                    Prob > F       =    0.0000

------------------------------------------------------------------------------
             |             Newey-West
           y |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
           x |   1.034833   .0312818    33.08   0.000     .9735074    1.096159
       _cons |   .0296797   .0341419     0.87   0.385    -.0372533    .0966128
------------------------------------------------------------------------------

. 

Looking forward for any help!

Thank you very much!

Sincerely,



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