xtabond2: explanatory variables being dropped because of collinearity

[jsaldain]

Dear Statalist readers,

I'm working with panel data, using the command xtabond2 for Arellano Bond estimator. When I run the command, most of my explanatory variables are dropped because of collinearity. To my knowledge, there is no collinearity between those variables; as a matter of fact, they only get dropped with one of three dependant variables.

In the attachment, you can find the results for the two dependant variables. Thank you in advance for any comments.

Kind regards,

Joaquín

[Nick Cox]

MS Word documents are not universally accessible. Many Stata users, even those using Windows, don't use it or have access to it. The best way to show results is to show them as if they were code, using the # button (toggle Advanced Editor settings).

[jsaldain]

Sorry, I didn't understand what you meant. How about .pdf?

[Nick Cox]

This is an example of output saved as if it were code.

1. Copy output from Stata.
2. Toggle Advanced Editor settings with A button.
3. Get Code mark-ups with # button.
4. Paste output between mark-ups.

. sysuse auto, clear
(1978 Automobile Data)

. tab rep78

     Repair |
Record 1978 |      Freq.     Percent        Cum.
------------+-----------------------------------
          1 |          2        2.90        2.90
          2 |          8       11.59       14.49
          3 |         30       43.48       57.97
          4 |         18       26.09       84.06
          5 |         11       15.94      100.00
------------+-----------------------------------
      Total |         69      100.00

Much easier to produce and to read than an attachment!

[jsaldain]

. xtabond2 desembolsos L.desembolsos T sucursal desempleo siga trim2-trim4 datos, noleveleq gmm(L.desembolsos) iv(T sucursal desemp
> leo siga trim2-trim4 datos)
Favoring speed over space. To switch, type or click on mata: mata set matafavor space, perm.
T dropped due to collinearity
sucursal dropped due to collinearity
desempleo dropped due to collinearity
trim2 dropped due to collinearity
trim4 dropped due to collinearity
datos dropped due to collinearity
Warning: Number of instruments may be large relative to number of observations.

Dynamic panel-data estimation, one-step difference GMM
------------------------------------------------------------------------------
Group variable: id                              Number of obs      =       286
Time variable : date                            Number of groups   =        11
Number of instruments = 231                     Obs per group: min =        26
Wald chi2(3)  =     35.51                                      avg =     26.00
Prob > chi2   =     0.000                                      max =        26
------------------------------------------------------------------------------
 desembolsos |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
 desembolsos |
         L1. |   .2991357   .0560448     5.34   0.000     .1892898    .4089816
             |
        siga |   17454.66   10802.47     1.62   0.106    -3717.786    38627.11
       trim3 |   128360.4   83157.56     1.54   0.123    -34625.43    291346.2
------------------------------------------------------------------------------
Instruments for first differences equation
  Standard
    D.(T sucursal desempleo siga trim2 trim3 trim4 datos)
  GMM-type (missing=0, separate instruments for each period unless collapsed)
    L(1/.).L.desembolsos
------------------------------------------------------------------------------
Arellano-Bond test for AR(1) in first differences: z =  -8.76  Pr > z =  0.000
Arellano-Bond test for AR(2) in first differences: z =  -1.34  Pr > z =  0.179
------------------------------------------------------------------------------
Sargan test of overid. restrictions: chi2(228)  = 287.18  Prob > chi2 =  0.005
  (Not robust, but not weakened by many instruments.)

Difference-in-Sargan tests of exogeneity of instrument subsets:
  gmm(L.desembolsos, lag(1 .))
    Sargan test excluding group:     chi2(4)    =   1.76  Prob > chi2 =  0.780
    Difference (null H = exogenous): chi2(224)  = 285.42  Prob > chi2 =  0.003
  iv(T sucursal desempleo siga trim2 trim3 trim4 datos)
    Sargan test excluding group:     chi2(221)  = 286.23  Prob > chi2 =  0.002
    Difference (null H = exogenous): chi2(7)    =   0.96  Prob > chi2 =  0.996

[jsaldain]

Thank you for the tip, Nick!