Bookmark and Share

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]

st: xtabond2 : dropping variables due to collinearity


From   john ebireri <[email protected]>
To   "[email protected]" <[email protected]>
Subject   st: xtabond2 : dropping variables due to collinearity
Date   Fri, 4 May 2012 05:10:11 -0700 (PDT)

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/


© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index