 |
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
st: twostep option in xtabond and xtabond2
Dear statalisters;
I'm trying to estimate a model using
TWOSTEP(xtabond)and I keep getting most of my
variables dropped(although there isn't any co
linearity problem))that is happened only when I add
the option (twostep); but it is OK when using the
one-step estimator.
My model as follows:
ROA(it)=ROA(i(t-1))+X(it)+ Z(t)+ year dummies+ uit
-X(it):changes over time per firm.
-Z(t): change only over time.
I used xtabond2 ..I got the same problem too. Also, I
used Stata-10 xtdpdsys..the same problem...I used the
system GMM I got the same problem of dropped
variables...
-I tried taking out the dummy variables or the
only-time-changing variables or both.. but .. no way
out?!
-Can any one suggest doing something please ?!
Here are my results:
The two step-results:
. xtabond ROA_new repo l.repo l_age l_ta l_sqASST
LaborPro l.LaborPro loans_ta l.loans_ta lq_ta
l.lq_ta oc_oi l.oc_oi mr
> 1 cpi RGDPgrw overhead HHI_loan stmktcap
FS_EF_TVTNIM yr2 yr3 yr4 yr10 yr14 yr15 , small
maxldep(1) maxlags
> (2) endogenous(eq_ta,lag(1,.)) two
small is a deprecated option
Arellano-Bond dynamic panel-data estimation Number of
obs = 150
Group variable: bank1 Number of
groups = 10
Time variable: year
Obs per
group: min = 15
avg = 15
max = 15
Number of instruments = 56 Wald
chi2(10) = 1115.06
Prob >
chi2 = 0.0000
Two-step results
------------------------------------------------------------------------------
ROA_new | Coef. Std. Err. z P>|z|
[95% Conf. Interval]
-------------+----------------------------------------------------------------
ROA_new |
L1. | (dropped)
eq_ta |
--. | (dropped)
L1. | (dropped)
repo |
--. | .0004397 .0003699 1.19 0.235
-.0002854 .0011647
L1. | -.0004259 .0003443 -1.24 0.216
-.0011007 .000249
l_age | (dropped)
l_ta | .0128983 .0046551 2.77 0.006
.0037744 .0220222
l_sqASST | -.0012276 .0004462 -2.75 0.006
-.0021021 -.0003531
LaborPro |
--. | .039226 .0106196 3.69 0.000
.0184119 .0600401
L1. | (dropped)
loans_ta |
--. | (dropped)
L1. | (dropped)
lq_ta |
--. | (dropped)
L1. | (dropped)
oc_oi |
--. | (dropped)
L1. | (dropped)
mr1 | (dropped)
cpi | .0002256 .0001576 1.43 0.152
-.0000833 .0005344
RGDPgrw | (dropped)
overhead | (dropped)
HHI_loan | (dropped)
stmktcap | (dropped)
FS_EF_TVTNIM | .0000362 .0000634 0.57 0.568
-.0000881 .0001606
yr2 | (dropped)
yr3 | .0003803 .0013689 0.28 0.781
-.0023026 .0030632
yr4 | (dropped)
yr10 | -.0007715 .001152 -0.67 0.503
-.0030293 .0014864
yr14 | (dropped)
yr15 | -.0021436 .001253 -1.71 0.087
-.0045994 .0003123
_cons | (dropped)
------------------------------------------------------------------------------
Warning: gmm two-step standard errors are biased;
robust standard
errors are recommended.
Instruments for differenced equation
GMM-type: L(2/2).ROA_new L(2/2).L.eq_ta
Standard: D.repo LD.repo D.l_age D.l_ta
D.l_sqASST D.LaborPro LD.LaborPro D.loans_ta
LD.loans_ta D.lq_ta LD.lq_ta
D.oc_oi LD.oc_oi D.mr1 D.cpi
D.RGDPgrw D.overhead D.HHI_loan D.stmktcap
D.FS_EF_TVTNIM D.yr2 D.yr3 D.yr4
D.yr10 D.yr14 D.yr15
Instruments for level equation
Standard: _cons
and the one-step results:
. xtabond ROA_new repo l.repo l_age l_ta l_sqASST
LaborPro l.LaborPro loans_ta l.loans_ta lq_ta
l.lq_ta oc_oi l.oc_oi mr
> 1 cpi RGDPgrw overhead HHI_loan stmktcap
FS_EF_TVTNIM yr2 yr3 yr4 yr10 yr14 yr15 , small
maxldep(1) maxlags
> (2) endogenous(eq_ta,lag(1,.)) vce(r)
small is a deprecated option
Arellano-Bond dynamic panel-data estimation Number of
obs = 150
Group variable: bank1 Number of
groups = 10
Time variable: year
Obs per
group: min = 15
avg = 15
max = 15
Number of instruments = 56 Wald
chi2(9) = 145.84
Prob >
chi2 = 0.0000
One-step results
------------------------------------------------------------------------------
| Robust
ROA_new | Coef. Std. Err. z P>|z|
[95% Conf. Interval]
-------------+----------------------------------------------------------------
ROA_new |
L1. | .2126414 .1077665 1.97 0.048
.0014229 .4238599
eq_ta |
--. | .1615146 .0429736 3.76 0.000
.077288 .2457413
L1. | -.1113438 .0473595 -2.35 0.019
-.2041667 -.0185209
repo |
--. | .0011525 .0008483 1.36 0.174
-.0005101 .0028151
L1. | .0033409 .0023355 1.43 0.153
-.0012366 .0079184
l_age | .004378 .0046463 0.94 0.346
-.0047286 .0134846
l_ta | .0827912 .0396641 2.09 0.037
.005051 .1605313
l_sqASST | -.004615 .001851 -2.49 0.013
-.0082429 -.0009871
LaborPro |
--. | .0264518 .0077233 3.42 0.001
.0113144 .0415891
L1. | -.0066753 .0075767 -0.88 0.378
-.0215254 .0081748
loans_ta |
--. | .0145953 .0138097 1.06 0.291
-.0124713 .0416619
L1. | -.0309462 .0116349 -2.66 0.008
-.0537502 -.0081422
lq_ta |
--. | -.0483015 .080012 -0.60 0.546
-.2051222 .1085192
L1. | -.0534446 .0921197 -0.58 0.562
-.2339959 .1271067
oc_oi |
--. | -.0161311 .001885 -8.56 0.000
-.0198257 -.0124366
L1. | -.0092007 .0043243 -2.13 0.033
-.0176762 -.0007252
mr1 | .0254298 .0561967 0.45 0.651
-.0847138 .1355734
cpi | .0014306 .0004817 2.97 0.003
.0004865 .0023748
RGDPgrw | -.0113802 .0114071 -1.00 0.318
-.0337377 .0109772
overhead | -.4649442 1.17125 -0.40 0.691
-2.760552 1.830664
HHI_loan | -.4583331 .2681471 -1.71 0.087
-.9838918 .0672256
stmktcap | -.0201806 .0559329 -0.36 0.718
-.129807 .0894459
FS_EF_TVTNIM | .0003301 .0006769 0.49 0.626
-.0009966 .0016568
yr2 | .0525664 .0312454 1.68 0.092
-.0086735 .1138063
yr3 | .0340745 .0205016 1.66 0.097
-.0061078 .0742568
yr4 | .006722 .0052519 1.28 0.201
-.0035716 .0170156
yr10 | .0042994 .0021829 1.97 0.049
.000021 .0085778
yr14 | -.0091958 .0047205 -1.95 0.051
-.0184478 .0000563
yr15 | -.0055886 .0036531 -1.53 0.126
-.0127486 .0015714
_cons | -.2958679 .2091643 -1.41 0.157
-.7058224 .1140866
------------------------------------------------------------------------------
Instruments for differenced equation
GMM-type: L(2/2).ROA_new L(2/2).L.eq_ta
Standard: D.repo LD.repo D.l_age D.l_ta
D.l_sqASST D.LaborPro LD.LaborPro D.loans_ta
LD.loans_ta D.lq_ta LD.lq_ta
D.oc_oi LD.oc_oi D.mr1 D.cpi
D.RGDPgrw D.overhead D.HHI_loan D.stmktcap
D.FS_EF_TVTNIM D.yr2 D.yr3 D.yr4
D.yr10 D.yr14 D.yr15
Instruments for level equation
Standard: _cons
Ahmed
____________________________________________________________________________________
Be a better friend, newshound, and
know-it-all with Yahoo! Mobile. Try it now. http://mobile.yahoo.com/;_ylt=Ahu06i62sR8HDtDypao8Wcj9tAcJ
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/