Statalist


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

Re: st: interpret xtabond one-step results


From   nicola.baldini2@unibo.it
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: interpret xtabond one-step results
Date   Sun, 12 Aug 2007 15:17:27 +0200

The following may help:
Roodman, D. 2006. How to Do xtabond2: An Introduction to "Difference" and "System" GMM in Stata. Working Paper 103. Center for Global Development, Washington.
Briefly:
1) I think that the related commands in Stata 10 may provide you the coefficients for the original variables, and not the differenced ones.
2) If Sargan test is significant, then the instruments may not be valid 
3) the autocorrelation tests say that there is AR(1) but not AR(2), and that it's safe to use lags 3 and deeper of y as instruments

Nicola

At 02.33 11/08/2007 -0400, "Jing Tong" wrote:
>Dear Stata users
>
>Could anyone please help me on the interpretation of the one-step results 
>as shown in the following example:
>
>1)      how to report the coefficients and Z-values of the independent 
>variables? I mean, instead of reporting as Wit and Wi(t-1) separately, 
>how to get the coefficient and Z-value for W itself? And how to tell the 
>significance level for W, if (P>|z|) for D1 is 0.000 (significant at 1% 
>level) while (P>|z|) for LD is 0.100 (significant at 10% level)? 
>2)      For Sargan test of over-identifying restrictions, is the 
>(prob>chi2) the probability of accepting the null hypotheses of valid 
>instruments? In this example, it seems to show the strong evidence 
>against the null. Does it mean the specification is not valid at all?
>3)      For AR(1) and AR(2) at the bottom of the output, is the (pr>z) 
>the probability of accepting the null hypotheses of no first-order and 
>second-order serial correlation? If the residuals are not serially 
>correlated, should there be evidence of significant negative first-order 
>serial correlation (i.e., pr>z close to 0) and not rejecting the null of 
>no second-order serial correlation (i.e., pr>z close to 1) ?
>
>
>. xtabond n l(0/1).w k , lags(2)
>
>Arellano-Bond dynamic panel data                Number of obs      
>=       611
>Group variable (i): id                          Number of groups   
>=       140
>
>                                                Wald chi2(5)       =    
>350.58
>
>Time variable (t): year                         min number of obs  
>=         4
>                                                max number of obs  
>=         6
>                                                mean number of obs =  
>4.364286
>
>One-step results
>- --------------------------------------------------------------------------
>- ----
>n            |      Coef.   Std. Err.      z    P>|z|     [95% Conf. 
>Interval]
>- -------------+------------------------------------------------------------
>- ----
>n            |
>          LD |   .3751428   .1050691     3.57   
>0.000     .1692112    .5810745
>         L2D |  -.0822723   .0422479    -1.95   0.051    -
>.1650766     .000532
>w            |
>          D1 |  -.4754038   .0564188    -8.43   0.000    -.5859825    -
>.364825
>          LD |    .208237   .0832401     2.50   
>0.012     .0450894    .3713847
>k            |
>          D1 |   .3802498   .0352074    10.80   
>0.000     .3112446     .449255
>_cons        |  -.0178497   .0041618    -4.29   0.000    -.0260068   -
>.0096926
>- --------------------------------------------------------------------------
>- ----
>Sargan test of over-identifying restrictions:     
>         chi2(25) =    97.07      Prob > chi2 = 0.0000
>
>Arellano-Bond test that average autocovariance in residuals of order 1 is 
>0:
>         H0: no autocorrelation   z =  -3.02   Pr > z = 0.0026
>Arellano-Bond test that average autocovariance in residuals of order 2 is 
>0:
>         H0: no autocorrelation   z =  -0.01   Pr > z = 0.9892
>
>
>Thank you very much for your help!
*
*   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