When i ran my panel IV estimation, using xtivreg2, I saw my instruments were very weak and thus tried to do a LIML estimation. However the results I am getting are exactly the same as the 2sls.
xtivreg2 pvexports corruptionindex democracy hipc lgdp lusd lnetoda (lcaid = lchiconstr), fe robust liml
Warning - singleton groups detected. 2 observation(s) not used.
FIXED EFFECTS ESTIMATION
------------------------
Number of groups = 37 Obs per group: min = 3
avg = 9.6
max = 14
LIML estimation
---------------
k =1.00000
lambda =1.00000
Estimates efficient for homoskedasticity only
Statistics robust to heteroskedasticity
Number of obs = 355
F( 7, 311) = 1.11
Prob > F = 0.3592
Total (centered) SS = 963983.0117 Centered R2 = -2.9281
Total (uncentered) SS = 963983.0117 Uncentered R2 = -2.9281
Residual SS = 3786656.654 Root MSE = 109.1
---------------------------------------------------------------------------------
| Robust
pvexports | Coef. Std. Err. z P>|z| [95% Conf. Interval]
----------------+----------------------------------------------------------------
lcaid | -53.34971 27.44603 -1.94 0.052 -107.1429 .4435234
corruptionindex | -2.051737 1.780379 -1.15 0.249 -5.541217 1.437742
democracy | 2.406781 12.67784 0.19 0.849 -22.44133 27.2549
hipc | -28.34563 23.23213 -1.22 0.222 -73.87976 17.18851
lgdp | 203.1531 131.84 1.54 0.123 -55.24857 461.5548
lusd | 3.682018 4.623068 0.80 0.426 -5.37903 12.74307
lnetoda | -7.031162 11.7945 -0.60 0.551 -30.14795 16.08563
---------------------------------------------------------------------------------
Underidentification test (Kleibergen-Paap rk LM statistic): 6.032
Chi-sq(1) P-val = 0.0140
------------------------------------------------------------------------------
Weak identification test (Cragg-Donald Wald F statistic): 4.261
(Kleibergen-Paap rk Wald F statistic): 5.184
Stock-Yogo weak ID test critical values: 10% maximal LIML size 16.38
15% maximal LIML size 8.96
20% maximal LIML size 6.66
25% maximal LIML size 5.53
Source: Stock-Yogo (2005). Reproduced by permission.
NB: Critical values are for Cragg-Donald F statistic and i.i.d. errors.
------------------------------------------------------------------------------
Hansen J statistic (overidentification test of all instruments): 0.000
(equation exactly identified)
------------------------------------------------------------------------------
Instrumented: lcaid
Included instruments: corruptionindex democracy hipc lgdp lusd lnetoda
Excluded instruments: lchiconstr
------------------------------------------------------------------------------
.
xtivreg2 pvexports corruptionindex democracy hipc lgdp lusd lnetoda (lcaid = lchiconstr), fe robust
Warning - singleton groups detected. 2 observation(s) not used.
FIXED EFFECTS ESTIMATION
------------------------
Number of groups = 37 Obs per group: min = 3
avg = 9.6
max = 14
IV (2SLS) estimation
--------------------
Estimates efficient for homoskedasticity only
Statistics robust to heteroskedasticity
Number of obs = 355
F( 7, 311) = 1.11
Prob > F = 0.3592
Total (centered) SS = 963983.0117 Centered R2 = -2.9281
Total (uncentered) SS = 963983.0117 Uncentered R2 = -2.9281
Residual SS = 3786656.654 Root MSE = 109.1
---------------------------------------------------------------------------------
| Robust
pvexports | Coef. Std. Err. z P>|z| [95% Conf. Interval]
----------------+----------------------------------------------------------------
lcaid | -53.34971 27.44603 -1.94 0.052 -107.1429 .4435234
corruptionindex | -2.051737 1.780379 -1.15 0.249 -5.541217 1.437742
democracy | 2.406781 12.67784 0.19 0.849 -22.44133 27.2549
hipc | -28.34563 23.23213 -1.22 0.222 -73.87976 17.18851
lgdp | 203.1531 131.84 1.54 0.123 -55.24857 461.5548
lusd | 3.682018 4.623068 0.80 0.426 -5.37903 12.74307
lnetoda | -7.031162 11.7945 -0.60 0.551 -30.14795 16.08563
---------------------------------------------------------------------------------
Underidentification test (Kleibergen-Paap rk LM statistic): 6.032
Chi-sq(1) P-val = 0.0140
------------------------------------------------------------------------------
Weak identification test (Cragg-Donald Wald F statistic): 4.261
(Kleibergen-Paap rk Wald F statistic): 5.184
Stock-Yogo weak ID test critical values: 10% maximal IV size 16.38
15% maximal IV size 8.96
20% maximal IV size 6.66
25% maximal IV size 5.53
Source: Stock-Yogo (2005). Reproduced by permission.
NB: Critical values are for Cragg-Donald F statistic and i.i.d. errors.
------------------------------------------------------------------------------
Hansen J statistic (overidentification test of all instruments): 0.000
(equation exactly identified)
------------------------------------------------------------------------------
Instrumented: lcaid
Included instruments: corruptionindex democracy hipc lgdp lusd lnetoda
Excluded instruments: lchiconstr
------------------------------------------------------------------------------
.
I don't understand why this is happening
xtivreg2 pvexports corruptionindex democracy hipc lgdp lusd lnetoda (lcaid = lchiconstr), fe robust liml
Warning - singleton groups detected. 2 observation(s) not used.
FIXED EFFECTS ESTIMATION
------------------------
Number of groups = 37 Obs per group: min = 3
avg = 9.6
max = 14
LIML estimation
---------------
k =1.00000
lambda =1.00000
Estimates efficient for homoskedasticity only
Statistics robust to heteroskedasticity
Number of obs = 355
F( 7, 311) = 1.11
Prob > F = 0.3592
Total (centered) SS = 963983.0117 Centered R2 = -2.9281
Total (uncentered) SS = 963983.0117 Uncentered R2 = -2.9281
Residual SS = 3786656.654 Root MSE = 109.1
---------------------------------------------------------------------------------
| Robust
pvexports | Coef. Std. Err. z P>|z| [95% Conf. Interval]
----------------+----------------------------------------------------------------
lcaid | -53.34971 27.44603 -1.94 0.052 -107.1429 .4435234
corruptionindex | -2.051737 1.780379 -1.15 0.249 -5.541217 1.437742
democracy | 2.406781 12.67784 0.19 0.849 -22.44133 27.2549
hipc | -28.34563 23.23213 -1.22 0.222 -73.87976 17.18851
lgdp | 203.1531 131.84 1.54 0.123 -55.24857 461.5548
lusd | 3.682018 4.623068 0.80 0.426 -5.37903 12.74307
lnetoda | -7.031162 11.7945 -0.60 0.551 -30.14795 16.08563
---------------------------------------------------------------------------------
Underidentification test (Kleibergen-Paap rk LM statistic): 6.032
Chi-sq(1) P-val = 0.0140
------------------------------------------------------------------------------
Weak identification test (Cragg-Donald Wald F statistic): 4.261
(Kleibergen-Paap rk Wald F statistic): 5.184
Stock-Yogo weak ID test critical values: 10% maximal LIML size 16.38
15% maximal LIML size 8.96
20% maximal LIML size 6.66
25% maximal LIML size 5.53
Source: Stock-Yogo (2005). Reproduced by permission.
NB: Critical values are for Cragg-Donald F statistic and i.i.d. errors.
------------------------------------------------------------------------------
Hansen J statistic (overidentification test of all instruments): 0.000
(equation exactly identified)
------------------------------------------------------------------------------
Instrumented: lcaid
Included instruments: corruptionindex democracy hipc lgdp lusd lnetoda
Excluded instruments: lchiconstr
------------------------------------------------------------------------------
.
xtivreg2 pvexports corruptionindex democracy hipc lgdp lusd lnetoda (lcaid = lchiconstr), fe robust
Warning - singleton groups detected. 2 observation(s) not used.
FIXED EFFECTS ESTIMATION
------------------------
Number of groups = 37 Obs per group: min = 3
avg = 9.6
max = 14
IV (2SLS) estimation
--------------------
Estimates efficient for homoskedasticity only
Statistics robust to heteroskedasticity
Number of obs = 355
F( 7, 311) = 1.11
Prob > F = 0.3592
Total (centered) SS = 963983.0117 Centered R2 = -2.9281
Total (uncentered) SS = 963983.0117 Uncentered R2 = -2.9281
Residual SS = 3786656.654 Root MSE = 109.1
---------------------------------------------------------------------------------
| Robust
pvexports | Coef. Std. Err. z P>|z| [95% Conf. Interval]
----------------+----------------------------------------------------------------
lcaid | -53.34971 27.44603 -1.94 0.052 -107.1429 .4435234
corruptionindex | -2.051737 1.780379 -1.15 0.249 -5.541217 1.437742
democracy | 2.406781 12.67784 0.19 0.849 -22.44133 27.2549
hipc | -28.34563 23.23213 -1.22 0.222 -73.87976 17.18851
lgdp | 203.1531 131.84 1.54 0.123 -55.24857 461.5548
lusd | 3.682018 4.623068 0.80 0.426 -5.37903 12.74307
lnetoda | -7.031162 11.7945 -0.60 0.551 -30.14795 16.08563
---------------------------------------------------------------------------------
Underidentification test (Kleibergen-Paap rk LM statistic): 6.032
Chi-sq(1) P-val = 0.0140
------------------------------------------------------------------------------
Weak identification test (Cragg-Donald Wald F statistic): 4.261
(Kleibergen-Paap rk Wald F statistic): 5.184
Stock-Yogo weak ID test critical values: 10% maximal IV size 16.38
15% maximal IV size 8.96
20% maximal IV size 6.66
25% maximal IV size 5.53
Source: Stock-Yogo (2005). Reproduced by permission.
NB: Critical values are for Cragg-Donald F statistic and i.i.d. errors.
------------------------------------------------------------------------------
Hansen J statistic (overidentification test of all instruments): 0.000
(equation exactly identified)
------------------------------------------------------------------------------
Instrumented: lcaid
Included instruments: corruptionindex democracy hipc lgdp lusd lnetoda
Excluded instruments: lchiconstr
------------------------------------------------------------------------------
.
I don't understand why this is happening
Comment