Dear all,

I'm new in this forum and the motivation for this post is that i ran a model with Gmm and I'm having a kind of situation that causes me some sort of concern.

I'm working with a dependent serie of manufacturing employment share in a regression over the gdppc and gdppc squared besides the population, N=41 T=6.
My concern is about the coefficients of the L1 and L2. To correct for the AR(2) I was obligated to inside the L2.empshare and when I made that the L1.empshare showed a coefficient over the unity although L1+L2<1. However, reading the paper of Blundell and Bond (1998) and Roodman (2009) and many others that i found, always they mention the coefficient alpha less than unity. I tried other database with N=96 T=5 but i obtained the same situation. All the tests are good, Hansen and AR(2).
I don't know if this blocks me to utilise the gmm or something like that.
If someone can help me i will really appreciate.
Thanks.

xtabond2 empshare L(1/2).empshare ln_gdppc ln_gdppc_2 ln_pop ln_pop_2 i.year2, ///
> gmmstyle (L(1/2).empshare ln_gdppc ln_gdppc_2, lag(1 5)collapse) ///
> iv(ln_pop ln_pop_2 i.year2) twostep robust
Favoring space over speed. To switch, type or click on mata: mata set matafavor speed, perm.
Warning: Two-step estimated covariance matrix of moments is singular.
Using a generalized inverse to calculate optimal weighting matrix for two-step estimation.
Difference-in-Sargan/Hansen statistics may be negative.

Dynamic panel-data estimation, two-step system GMM
------------------------------------------------------------------------------
Group variable: countryid Number of obs = 164
Time variable : year2 Number of groups = 41
Number of instruments = 24 Obs per group: min = 4
Wald chi2(12) = 866.09 avg = 4.00
Prob > chi2 = 0.000 max = 4
------------------------------------------------------------------------------
| Corrected
empshare | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
empshare |
L1. | 1.275678 .1732141 7.36 0.000 .9361851 1.615172
L2. | -.4880215 .1056025 -4.62 0.000 -.6949985 -.2810444
|
ln_gdppc | .0854323 .1366405 0.63 0.532 -.1823781 .3532427
ln_gdppc_2 | -.0044043 .0071298 -0.62 0.537 -.0183786 .0095699
ln_pop | .00209 .0036733 0.57 0.569 -.0051097 .0092896
ln_pop_2 | -.0001732 .0004833 -0.36 0.720 -.0011204 .000774
|
year2 |
1 | 0 (empty)
2 | 0 (omitted)
3 | 0 (omitted)
4 | -.0105388 .0037834 -2.79 0.005 -.0179541 -.0031235
5 | -.0082161 .0042942 -1.91 0.056 -.0166326 .0002003
6 | -.0117497 .0050857 -2.31 0.021 -.0217175 -.0017818
|
_cons | -.3798522 .6330268 -0.60 0.548 -1.620562 .8608575
------------------------------------------------------------------------------
Instruments for first differences equation
Standard
D.(ln_pop ln_pop_2 1b.year2 2.year2 3.year2 4.year2 5.year2 6.year2)
GMM-type (missing=0, separate instruments for each period unless collapsed)
L(1/5).(L.empshare L2.empshare ln_gdppc ln_gdppc_2) collapsed
Instruments for levels equation
Standard
ln_pop ln_pop_2 1b.year2 2.year2 3.year2 4.year2 5.year2 6.year2
_cons
GMM-type (missing=0, separate instruments for each period unless collapsed)
D.(L.empshare L2.empshare ln_gdppc ln_gdppc_2) collapsed
------------------------------------------------------------------------------
Arellano-Bond test for AR(1) in first differences: z = -2.19 Pr > z = 0.029
Arellano-Bond test for AR(2) in first differences: z = -0.08 Pr > z = 0.934
------------------------------------------------------------------------------
Sargan test of overid. restrictions: chi2(11) = 23.19 Prob > chi2 = 0.017
(Not robust, but not weakened by many instruments.)
Hansen test of overid. restrictions: chi2(11) = 15.65 Prob > chi2 = 0.155
(Robust, but weakened by many instruments.)

Difference-in-Hansen tests of exogeneity of instrument subsets:
GMM instruments for levels
Hansen test excluding group: chi2(7) = 14.54 Prob > chi2 = 0.042
Difference (null H = exogenous): chi2(4) = 1.11 Prob > chi2 = 0.892
iv(ln_pop ln_pop_2 1b.year2 2.year2 3.year2 4.year2 5.year2 6.year2)
Hansen test excluding group: chi2(6) = 13.36 Prob > chi2 = 0.038
Difference (null H = exogenous): chi2(5) = 2.29 Prob > chi2 = 0.807