Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: st: xtabond2 : dropping variables due to collinearity
From
Muhammad Anees <[email protected]>
To
[email protected]
Subject
Re: st: xtabond2 : dropping variables due to collinearity
Date
Fri, 4 May 2012 17:26:50 +0500
Hi,
What if you do not add gdp and/or use gdp with variables only not with
the variables/gdp. Is this not contradictory?
I think you only need either to use the level variables or scale it on
gdp, so when scaling over gdp, you might not need to use gdp in the
regression. Both will work fine.
More advice on the issues would also be of interest to me. Let's see
what comes up.
Thanking you.
Anees
On Fri, May 4, 2012 at 5:10 PM, john ebireri <[email protected]> wrote:
> Dear stalist users,
>
> I have this problem where stata drops some of my variables due to collinearity when using xtabond2.
>
> The problem occurs when I introduce gdp to the model. The variables that get dropped due to collinearity are the lagged dependent variable-manufacturing value added/gdp, extrade-manufactured exports/gdp and inv-gross domestic investment/gdp. Any suggestions or solution please? I will show the results of both models:
>
>
> . xtabond2 manvaad l.manvaad bcon fbank gbank bnkdev extrade govexp inv, gmmstyle(l.manvaad bcon fbank gbank, laglimits(2 1) collapse) ivstyle(bnkdev extrade govexp inv, equation(level)) robust twostep
> Favoring space over speed. To switch, type or click on mata: mata set matafavor speed, perm.
>
> Dynamic panel-data estimation, two-step system GMM
> ------------------------------------------------------------------------------
> Group variable: id Number of obs = 75
> Time variable : year Number of groups = 20
> Number of instruments = 17 Obs per group: min = 3
> Wald chi2(8) = 1716.17 avg = 3.75
> Prob > chi2 = 0.000 max = 5
> ------------------------------------------------------------------------------
> | Corrected
> manvaad | Coef. Std. Err. z P>|z| [95% Conf. Interval]
> -------------+----------------------------------------------------------------
> manvaad |
> L1. | .8575447 .1152912 7.44 0.000 .6315781 1.083511
> |
> bcon | .0567681 .0241147 2.35 0.019 .0095042 .104032
> fbank | .0132493 .019177 0.69 0.490 -.0243368 .0508355
> gbank | .0170984 .0090695 1.89 0.059 -.0006776 .0348743
> bnkdev | .0328068 .0139792 2.35 0.019 .005408 .0602055
> extrade | .0669047 .0152983 4.37 0.000 .0369205 .0968889
> govexp | .0025887 .0003329 7.78 0.000 .0019363 .0032411
> inv | .0990203 .0552824 1.79 0.073 -.0093312 .2073717
> _cons | -.0693288 .0325576 -2.13 0.033 -.1331405 -.0055171
> ------------------------------------------------------------------------------
> Instruments for first differences equation
> GMM-type (missing=0, separate instruments for each period unless collapsed)
> L(1/2).(L.manvaad bcon fbank gbank) collapsed
> Instruments for levels equation
> Standard
> _cons
> bnkdev extrade govexp inv
> GMM-type (missing=0, separate instruments for each period unless collapsed)
> D.(L.manvaad bcon fbank gbank) collapsed
> ------------------------------------------------------------------------------
> Arellano-Bond test for AR(1) in first differences: z = -1.16 Pr > z = 0.245
> Arellano-Bond test for AR(2) in first differences: z = -1.35 Pr > z = 0.176
> ------------------------------------------------------------------------------
> Sargan test of overid. restrictions: chi2(8) = 5.17 Prob > chi2 = 0.740
> (Not robust, but not weakened by many instruments.)
> Hansen test of overid. restrictions: chi2(8) = 5.02 Prob > chi2 = 0.755
> (Robust, but can be weakened by many instruments.)
>
> Difference-in-Hansen tests of exogeneity of instrument subsets:
> GMM instruments for levels
> Hansen test excluding group: chi2(4) = 1.88 Prob > chi2 = 0.757
> Difference (null H = exogenous): chi2(4) = 3.14 Prob > chi2 = 0.535
> iv(bnkdev extrade govexp inv, eq(level))
> Hansen test excluding group: chi2(4) = 3.80 Prob > chi2 = 0.434
> Difference (null H = exogenous): chi2(4) = 1.22 Prob > chi2 = 0.874
>
> The second estimation has got gdp included in it:
>
> . xtabond2 manvaad l.manvaad bcon fbank gbank bnkdev extrade govexp inv gdp, gmmstyle(l.manvaad bcon fbank g
>> bank, laglimits(2 1) collapse) ivstyle(bnkdev extrade govexp inv gdp, equation(level)) robust twostep
> Favoring speed over space. To switch, type or click on mata: mata set matafavor space, perm.
> L.manvaad dropped due to collinearity
> extrade dropped due to collinearity
> inv dropped due to collinearity
>
> Dynamic panel-data estimation, two-step system GMM
> ------------------------------------------------------------------------------
> Group variable: id Number of obs = 75
> Time variable : year Number of groups = 20
> Number of instruments = 18 Obs per group: min = 3
> Wald chi2(4) = 11.25 avg = 3.75
> Prob > chi2 = 0.024 max = 5
> ------------------------------------------------------------------------------
> | Corrected
> manvaad | Coef. Std. Err. z P>|z| [95% Conf. Interval]
> -------------+----------------------------------------------------------------
> bcon | .1790888 .0855126 2.09 0.036 .0114872 .3466903
> fbank | -.0487292 .0512504 -0.95 0.342 -.149178 .0517197
> gbank | -.0265053 .0343287 -0.77 0.440 -.0937883 .0407777
> bnkdev | .1231247 .0603747 2.04 0.041 .0047924 .2414571
> govexp | .0043526 .0014221 3.06 0.002 .0015654 .0071399
> gdp | 8.00e-14 7.33e-14 1.09 0.275 -6.37e-14 2.24e-13
> _cons | .0227216 .0973777 0.23 0.816 -.1681351 .2135783
> ------------------------------------------------------------------------------
> Instruments for first differences equation
> GMM-type (missing=0, separate instruments for each period unless collapsed)
> L(1/2).(L.manvaad bcon fbank gbank) collapsed
> Instruments for levels equation
> Standard
> _cons
> bnkdev extrade govexp inv gdp
> GMM-type (missing=0, separate instruments for each period unless collapsed)
> D.(L.manvaad bcon fbank gbank) collapsed
> ------------------------------------------------------------------------------
> Arellano-Bond test for AR(1) in first differences: z = 1.38 Pr > z = 0.167
> Arellano-Bond test for AR(2) in first differences: z = -0.24 Pr > z = 0.813
> ------------------------------------------------------------------------------
> Sargan test of overid. restrictions: chi2(11) = 81.53 Prob > chi2 = 0.000
> (Not robust, but not weakened by many instruments.)
> Hansen test of overid. restrictions: chi2(11) = 16.60 Prob > chi2 = 0.120
> (Robust, but can be weakened by many instruments.)
>
> Difference-in-Hansen tests of exogeneity of instrument subsets:
> GMM instruments for levels
> Hansen test excluding group: chi2(7) = 11.34 Prob > chi2 = 0.125
> Difference (null H = exogenous): chi2(4) = 5.26 Prob > chi2 = 0.262
> iv(bnkdev extrade govexp inv gdp, eq(level))
> Hansen test excluding group: chi2(6) = 7.64 Prob > chi2 = 0.266
> Difference (null H = exogenous): chi2(5) = 8.95 Prob > chi2 = 0.111
>
> .
> Thanks,
>
> John.
>
> *
> * For searches and help try:
> * http://www.stata.com/help.cgi?search
> * http://www.stata.com/support/statalist/faq
> * http://www.ats.ucla.edu/stat/stata/
--
Best
---------------------------
Muhammad Anees
Assistant Professor/Programme Coordinator
COMSATS Institute of Information Technology
Attock 43600, Pakistan
http://www.aneconomist.com
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/