st: Persistency measurement on unbalanced panel (dynamic panel regression)

 From "Christian Stammschulte" To Subject st: Persistency measurement on unbalanced panel (dynamic panel regression) Date Sun, 7 Feb 2010 10:28:28 +0100

```Dear Stata Users,

I am new to STATA and I am currently trying to investigate profit
persistency patterns.
Given an unbalanced panel, short time periods and 3,000 companies for
investigation with demeaned RoEs I was trying to estimate persistency by
applying xtabond2 with three lags, twostep, robust and orthogonal approach.

However, two questions on this:

(1) I am receiving the error message "Two-step estimated covariance matrix
of moments is singular." Could you briefly explain to me why this is the
case?
(2) My Sargan and Hansen test are showing p-values of 0.0000. Since my
dataset only comprises RoEs/ lags in the model I assume this is the reason.
However, since I am only interested in persistency measurement I start
questioning whether the given procedure it is the right way to do it or
whether I am missing anything?

You will find the STATA output below.

Any help/ thoughts would be much appreciated.

Best regards and many thanks

Christian

*** STATA Output ***

. xtabond2  demeaned_win99_roae_ L.demeaned_win99_roae_
L2.demeaned_win99_roae_ L3.demeaned_win99_roae_,
gmmstyle(L.demeaned_win99_roae_) twostep robust orthogonal

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: id                              Number of obs      =
25084
Time variable : year                            Number of groups   =
3375
Number of instruments = 87                      Obs per group: min =
1
Wald chi2(3)  =   1103.60                                      avg =
7.43
Prob > chi2   =     0.000                                      max =
11
----------------------------------------------------------------------------
--
|              Corrected
demeaned_w~_ |      Coef.   Std. Err.      z    P>|z|     [95% Conf.
Interval]
-------------+--------------------------------------------------------------
--
demeaned_w~_ |
L1. |    .539917   .0177421    30.43   0.000      .505143
.5746909
L2. |   .1872978   .0155508    12.04   0.000     .1568188
.2177768
L3. |   .0594477   .0123591     4.81   0.000     .0352244
.0836711
|
_cons |  -.0004252   .0003175    -1.34   0.180    -.0010475
.0001971
----------------------------------------------------------------------------
--
Instruments for orthogonal deviations equation
GMM-type (missing=0, separate instruments for each period unless
collapsed)
L(1/.).L.demeaned_win99_roae_
Instruments for levels equation
Standard
_cons
GMM-type (missing=0, separate instruments for each period unless
collapsed)
D.L.demeaned_win99_roae_
----------------------------------------------------------------------------
--
Arellano-Bond test for AR(1) in first differences: z = -18.61  Pr > z =
0.000
Arellano-Bond test for AR(2) in first differences: z =  -0.59  Pr > z =
0.557
----------------------------------------------------------------------------
--
Sargan test of overid. restrictions: chi2(83)   = 624.28  Prob > chi2 =
0.000
(Not robust, but not weakened by many instruments.)
Hansen test of overid. restrictions: chi2(83)   = 248.87  Prob > chi2 =
0.000
(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(72)   = 214.50  Prob > chi2 =
0.000
Difference (null H = exogenous): chi2(11)   =  34.37  Prob > chi2 =
0.000

*
*   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/
```