Verbeek 5ed. Chapter 10 - Panel Data
Examples
----------------------------------------------------------------------------------------------------
name: SN
log: \5iexample10_s.smcl
log type: smcl
opened on: 5 Jun 2020, 01:48:00
. **********************************************
. * Solomon Negash - Examples
. * Verbeek(2017). A Giude To Modern Econometrics. 5ed.
. * STATA Program, version 16.1.
. * Chapter 10 - Models Based On Panel Data
. ******************** **** *********************
. * Table 10.2 Estimation results wage equation, males 1980-1987
. use "Data\males.dta", clear
. xtset NR YEAR
panel variable: NR (strongly balanced)
time variable: YEAR, 1980 to 1987
delta: 1 unit
. eststo BE: xtreg WAGE SCHOOL EXPER EXPER2 UNION MAR BLACK HISP PUB, be
Between regression (regression on group means) Number of obs = 4,360
Group variable: NR Number of groups = 545
R-sq: Obs per group:
within = 0.0470 min = 8
between = 0.2196 avg = 8.0
overall = 0.1371 max = 8
F(8,536) = 18.85
sd(u_i + avg(e_i.))= .3477522 Prob > F = 0.0000
------------------------------------------------------------------------------
WAGE | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
SCHOOL | .0947911 .0109178 8.68 0.000 .0733442 .1162381
EXPER | -.0502077 .0503689 -1.00 0.319 -.1491524 .048737
EXPER2 | .0051068 .0032142 1.59 0.113 -.0012071 .0114208
UNION | .2743194 .0471273 5.82 0.000 .1817426 .3668963
MAR | .1445897 .0412654 3.50 0.000 .063528 .2256515
BLACK | -.1391368 .0489084 -2.84 0.005 -.2352124 -.0430612
HISP | .0054832 .0427436 0.13 0.898 -.0784823 .0894488
PUB | -.0563215 .1090691 -0.52 0.606 -.2705768 .1579337
_cons | .4903902 .2211917 2.22 0.027 .0558814 .924899
------------------------------------------------------------------------------
. eststo FE: xtreg WAGE SCHOOL EXPER EXPER2 UNION MAR BLACK HISP PUB, fe
note: SCHOOL omitted because of collinearity
note: BLACK omitted because of collinearity
note: HISP omitted because of collinearity
Fixed-effects (within) regression Number of obs = 4,360
Group variable: NR Number of groups = 545
R-sq: Obs per group:
within = 0.1782 min = 8
between = 0.0006 avg = 8.0
overall = 0.0642 max = 8
F(5,3810) = 165.26
corr(u_i, Xb) = -0.1130 Prob > F = 0.0000
------------------------------------------------------------------------------
WAGE | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
SCHOOL | 0 (omitted)
EXPER | .116457 .0084309 13.81 0.000 .0999275 .1329865
EXPER2 | -.0042886 .0006054 -7.08 0.000 -.0054756 -.0031015
UNION | .081203 .0193159 4.20 0.000 .0433325 .1190736
MAR | .0451061 .0183114 2.46 0.014 .009205 .0810072
BLACK | 0 (omitted)
HISP | 0 (omitted)
PUB | .0349267 .0386082 0.90 0.366 -.040768 .1106214
_cons | 1.065698 .0266766 39.95 0.000 1.013396 1.118
-------------+----------------------------------------------------------------
sigma_u | .39989822
sigma_e | .35126372
rho | .56447541 (fraction of variance due to u_i)
------------------------------------------------------------------------------
F test that all u_i=0: F(544, 3810) = 9.71 Prob > F = 0.0000
. eststo POLS: reg WAGE SCHOOL EXPER EXPER2 UNION MAR BLACK HISP PUB,
Source | SS df MS Number of obs = 4,360
-------------+---------------------------------- F(8, 4351) = 124.76
Model | 230.721836 8 28.8402295 Prob > F = 0.0000
Residual | 1005.80781 4,351 .231167043 R-squared = 0.1866
-------------+---------------------------------- Adj R-squared = 0.1851
Total | 1236.52964 4,359 .283672779 Root MSE = .4808
------------------------------------------------------------------------------
WAGE | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
SCHOOL | .0993678 .0046829 21.22 0.000 .090187 .1085487
EXPER | .089138 .0101215 8.81 0.000 .0692948 .1089813
EXPER2 | -.0028468 .0007077 -4.02 0.000 -.0042343 -.0014594
UNION | .1799043 .0172146 10.45 0.000 .1461549 .2136537
MAR | .1076212 .0157053 6.85 0.000 .0768308 .1384115
BLACK | -.1438227 .023563 -6.10 0.000 -.1900182 -.0976271
HISP | .0156503 .0208197 0.75 0.452 -.0251668 .0564674
PUB | .0035461 .037474 0.09 0.925 -.0699219 .0770142
_cons | -.0343724 .0646723 -0.53 0.595 -.1611631 .0924182
------------------------------------------------------------------------------
. eststo RE: xtreg WAGE SCHOOL EXPER EXPER2 UNION MAR BLACK HISP PUB, re
Random-effects GLS regression Number of obs = 4,360
Group variable: NR Number of groups = 545
R-sq: Obs per group:
within = 0.1776 min = 8
between = 0.1835 avg = 8.0
overall = 0.1808 max = 8
Wald chi2(8) = 944.56
corr(u_i, X) = 0 (assumed) Prob > chi2 = 0.0000
------------------------------------------------------------------------------
WAGE | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
SCHOOL | .1010237 .0089219 11.32 0.000 .0835372 .1185103
EXPER | .1117851 .0082709 13.52 0.000 .0955744 .1279959
EXPER2 | -.0040575 .000592 -6.85 0.000 -.0052177 -.0028972
UNION | .1064134 .0178669 5.96 0.000 .0713949 .1414319
MAR | .0625465 .0167762 3.73 0.000 .0296658 .0954272
BLACK | -.1440026 .0476439 -3.02 0.003 -.237383 -.0506223
HISP | .0197269 .0426303 0.46 0.644 -.0638269 .1032807
PUB | .0301555 .0364671 0.83 0.408 -.0413187 .1016296
_cons | -.1043113 .110834 -0.94 0.347 -.3215421 .1129194
-------------+----------------------------------------------------------------
sigma_u | .32482045
sigma_e | .35126372
rho | .46094736 (fraction of variance due to u_i)
------------------------------------------------------------------------------
. estout BE FE POLS RE, cells(b(nostar fmt(3)) se(par fmt(3))) stats(r2 r2_p N, fmt(%5.0g %5.0g) lab
els(R-Squared Psuedo_R-Sqaured N )) varlabels(_cons constant) varwidth(10) ti("Table 10.2 Estimati
on results wage equation, males 1980-1987")
Table 10.2 Estimation results wage equation, males 1980-1987
--------------------------------------------------------------
BE FE POLS RE
b/se b/se b/se b/se
--------------------------------------------------------------
SCHOOL 0.095 0.000 0.099 0.101
(0.011) (.) (0.005) (0.009)
EXPER -0.050 0.116 0.089 0.112
(0.050) (0.008) (0.010) (0.008)
EXPER2 0.005 -0.004 -0.003 -0.004
(0.003) (0.001) (0.001) (0.001)
UNION 0.274 0.081 0.180 0.106
(0.047) (0.019) (0.017) (0.018)
MAR 0.145 0.045 0.108 0.063
(0.041) (0.018) (0.016) (0.017)
BLACK -0.139 0.000 -0.144 -0.144
(0.049) (.) (0.024) (0.048)
HISP 0.005 0.000 0.016 0.020
(0.043) (.) (0.021) (0.043)
PUB -0.056 0.035 0.004 0.030
(0.109) (0.039) (0.037) (0.036)
constant 0.490 1.066 -0.034 -0.104
(0.221) (0.027) (0.065) (0.111)
--------------------------------------------------------------
R-Squared .22 .178 .187
Psuedo_R~d
N 4.4e+03 4.4e+03 4.4e+03 4.4e+03
--------------------------------------------------------------
. * Table 10.3 OLS, within and OLS-FD estimation results dynamic model
. use "Data\debtratio.dta", clear
. xtset gvkey yeara
panel variable: gvkey (unbalanced)
time variable: yeara, 1986 to 2001, but with gaps
delta: 1 unit
. eststo POLS: reg mdr l.mdr lagebit_ta lagmb lagdep_ta laglnta lagfa_ta lagrd_dum lagrd_ta lagindme
dian lagrated, r
Linear regression Number of obs = 19,573
F(10, 19562) = 5490.12
Prob > F = 0.0000
R-squared = 0.7409
Root MSE = .12526
------------------------------------------------------------------------------
| Robust
mdr | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
mdr |
L1. | .8835036 .0050071 176.45 0.000 .8736893 .8933179
|
lagebit_ta | -.0323378 .0066229 -4.88 0.000 -.0453192 -.0193563
lagmb | .0016432 .0006228 2.64 0.008 .0004225 .0028639
lagdep_ta | -.2605179 .03579 -7.28 0.000 -.3306693 -.1903666
laglnta | -.0006704 .0005836 -1.15 0.251 -.0018143 .0004735
lagfa_ta | .0201215 .0054843 3.67 0.000 .0093718 .0308711
lagrd_dum | .0068896 .0020858 3.30 0.001 .0028013 .0109778
lagrd_ta | -.1202051 .0131558 -9.14 0.000 -.1459915 -.0944187
lagindmedian | .0321225 .0096277 3.34 0.001 .0132514 .0509935
lagrated | .0071341 .0027677 2.58 0.010 .0017092 .0125589
_cons | .0581818 .010525 5.53 0.000 .0375518 .0788117
------------------------------------------------------------------------------
. eststo FE: xtreg mdr l.mdr lagebit_ta lagmb lagdep_ta laglnta lagfa_ta lagrd_dum lagrd_ta lagindme
dian lagrated, fe r
Fixed-effects (within) regression Number of obs = 19,573
Group variable: gvkey Number of groups = 3,777
R-sq: Obs per group:
within = 0.3404 min = 1
between = 0.6409 avg = 5.2
overall = 0.5623 max = 15
F(10,3776) = 322.30
corr(u_i, Xb) = 0.0823 Prob > F = 0.0000
(Std. Err. adjusted for 3,777 clusters in gvkey)
------------------------------------------------------------------------------
| Robust
mdr | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
mdr |
L1. | .5349825 .011908 44.93 0.000 .5116358 .5583293
|
lagebit_ta | -.0500329 .0111017 -4.51 0.000 -.0717989 -.028267
lagmb | .0022776 .0010086 2.26 0.024 .0003 .0042551
lagdep_ta | -.1239544 .0709401 -1.75 0.081 -.263039 .0151301
laglnta | .0380301 .0030688 12.39 0.000 .0320135 .0440468
lagfa_ta | .0593436 .0170793 3.47 0.001 .025858 .0928291
lagrd_dum | .0000598 .0080767 0.01 0.994 -.0157753 .0158948
lagrd_ta | -.0656762 .0264011 -2.49 0.013 -.1174379 -.0139145
lagindmedian | .1672179 .022364 7.48 0.000 .1233712 .2110647
lagrated | .0205898 .0058294 3.53 0.000 .0091607 .032019
_cons | -.6008348 .0569212 -10.56 0.000 -.712434 -.4892356
-------------+----------------------------------------------------------------
sigma_u | .14361663
sigma_e | .11333049
rho | .61625402 (fraction of variance due to u_i)
------------------------------------------------------------------------------
. eststo FD: reg d.mdr l.d.mdr d.lagebit_ta d.lagmb d.lagdep_ta d.laglnta d.lagfa_ta d.lagrd_dum d.l
agrd_ta d.lagindmedian d.lagrated, r nocons
Linear regression Number of obs = 15,039
F(10, 15029) = 34.43
Prob > F = 0.0000
R-squared = 0.0325
Root MSE = .12436
------------------------------------------------------------------------------
| Robust
D.mdr | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
mdr |
LD. | -.110409 .0124587 -8.86 0.000 -.1348296 -.0859884
|
lagebit_ta |
D1. | -.0460275 .0101405 -4.54 0.000 -.0659041 -.0261508
|
lagmb |
D1. | .0027553 .0010367 2.66 0.008 .0007232 .0047873
|
lagdep_ta |
D1. | .1837711 .0797316 2.30 0.021 .0274874 .3400548
|
laglnta |
D1. | .0727863 .0047634 15.28 0.000 .0634494 .0821231
|
lagfa_ta |
D1. | .1012892 .0179843 5.63 0.000 .0660378 .1365407
|
lagrd_dum |
D1. | -.0173716 .0090703 -1.92 0.055 -.0351504 .0004073
|
lagrd_ta |
D1. | -.0516802 .0286951 -1.80 0.072 -.1079261 .0045657
|
lagindmedian |
D1. | .1787605 .0256968 6.96 0.000 .1283917 .2291294
|
lagrated |
D1. | .0114568 .0065153 1.76 0.079 -.001314 .0242276
------------------------------------------------------------------------------
. * Table 10.4 IV and GMM estimation results dynamic model
. u "Data/debtratio.dta", clear
. eststo AH1: ivreg d.mdr (l.d.mdr=l2.d.mdr) d.lagebit_ta d.lagmb d.lagdep_ta d.laglnta d.lagfa_ta d
.lagrd_dum d.lagrd_ta d.lagindmedian d.lagrated, nocons r
Instrumental variables (2SLS) regression Number of obs = 11,732
F(10, 11722) = 0.27
Prob > F = 0.9878
R-squared = .
Root MSE = .92404
------------------------------------------------------------------------------
| Robust
D.mdr | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
mdr |
LD. | 8.555149 11.3126 0.76 0.450 -13.61943 30.72973
|
lagebit_ta |
D1. | 1.480619 2.008451 0.74 0.461 -2.456278 5.417517
|
lagmb |
D1. | .2955019 .3809864 0.78 0.438 -.4512949 1.042299
|
lagdep_ta |
D1. | -2.439425 3.461122 -0.70 0.481 -9.223799 4.344949
|
laglnta |
D1. | -.6685727 .9734362 -0.69 0.492 -2.57667 1.239524
|
lagfa_ta |
D1. | -1.337006 1.887529 -0.71 0.479 -5.036877 2.362866
|
lagrd_dum |
D1. | -.023188 .0957618 -0.24 0.809 -.2108971 .1645211
|
lagrd_ta |
D1. | 1.068151 1.512725 0.71 0.480 -1.897042 4.033344
|
lagindmedian |
D1. | -4.118354 5.618381 -0.73 0.464 -15.13132 6.894608
|
lagrated |
D1. | -.3376334 .4584422 -0.74 0.461 -1.236256 .5609896
------------------------------------------------------------------------------
Instrumented: LD.mdr
Instruments: D.lagebit_ta D.lagmb D.lagdep_ta D.laglnta D.lagfa_ta
D.lagrd_dum D.lagrd_ta D.lagindmedian D.lagrated L2D.mdr
------------------------------------------------------------------------------
. eststo AH2: ivreg d.mdr (l.d.mdr=l2.mdr) d.lagebit_ta d.lagmb d.lagdep_ta d.laglnta d.lagfa_ta d.l
agrd_dum d.lagrd_ta d.lagindmedian d.lagrated, nocons r
Instrumental variables (2SLS) regression Number of obs = 15,039
F(10, 15029) = 10.02
Prob > F = 0.0000
R-squared = .
Root MSE = .18243
------------------------------------------------------------------------------
| Robust
D.mdr | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
mdr |
LD. | 1.124607 .3770692 2.98 0.003 .3855051 1.863708
|
lagebit_ta |
D1. | .1629408 .0655393 2.49 0.013 .0344758 .2914059
|
lagmb |
D1. | .0397811 .0114787 3.47 0.001 .0172814 .0622808
|
lagdep_ta |
D1. | -.1506395 .1625619 -0.93 0.354 -.4692806 .1680016
|
laglnta |
D1. | -.0319194 .0325089 -0.98 0.326 -.0956408 .031802
|
lagfa_ta |
D1. | -.1239245 .073548 -1.68 0.092 -.2680874 .0202385
|
lagrd_dum |
D1. | -.0206255 .0144243 -1.43 0.153 -.048899 .0076479
|
lagrd_ta |
D1. | .0992718 .061007 1.63 0.104 -.0203094 .218853
|
lagindmedian |
D1. | -.4627651 .1995744 -2.32 0.020 -.8539552 -.0715749
|
lagrated |
D1. | -.0419514 .0191066 -2.20 0.028 -.0794027 -.0045001
------------------------------------------------------------------------------
Instrumented: LD.mdr
Instruments: D.lagebit_ta D.lagmb D.lagdep_ta D.laglnta D.lagfa_ta
D.lagrd_dum D.lagrd_ta D.lagindmedian D.lagrated L2.mdr
------------------------------------------------------------------------------
. eststo AB1: xtabond mdr lagebit_ta lagmb lagdep_ta laglnta lagfa_ta lagrd_dum lagrd_ta lagindmedi
an lagrated, nocons
Arellano-Bond dynamic panel-data estimation Number of obs = 15,039
Group variable: gvkey Number of groups = 2,996
Time variable: yeara
Obs per group:
min = 1
avg = 5.019693
max = 14
Number of instruments = 114 Wald chi2(10) = 792.24
Prob > chi2 = 0.0000
One-step results
------------------------------------------------------------------------------
mdr | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
mdr |
L1. | .4716 .0367014 12.85 0.000 .3996665 .5435335
|
lagebit_ta | .0501892 .0112482 4.46 0.000 .0281433 .0722352
lagmb | .0210889 .0017668 11.94 0.000 .0176259 .0245518
lagdep_ta | -.0381627 .0769418 -0.50 0.620 -.1889658 .1126404
laglnta | .0253849 .0046089 5.51 0.000 .0163516 .0344183
lagfa_ta | -.0049994 .0194774 -0.26 0.797 -.0431745 .0331756
lagrd_dum | -.0183677 .0089972 -2.04 0.041 -.036002 -.0007335
lagrd_ta | .0186554 .0334981 0.56 0.578 -.0469996 .0843103
lagindmedian | .1012676 .0335516 3.02 0.003 .0355078 .1670275
lagrated | -.0089812 .0072924 -1.23 0.218 -.0232741 .0053116
------------------------------------------------------------------------------
Instruments for differenced equation
GMM-type: L(2/.).mdr
Standard: D.lagebit_ta D.lagmb D.lagdep_ta D.laglnta D.lagfa_ta
D.lagrd_dum D.lagrd_ta D.lagindmedian D.lagrated
. eststo AB2: xtabond mdr lagebit_ta lagmb lagdep_ta laglnta lagfa_ta lagrd_dum lagrd_ta lagindmedi
an lagrated, nocons two
Arellano-Bond dynamic panel-data estimation Number of obs = 15,039
Group variable: gvkey Number of groups = 2,996
Time variable: yeara
Obs per group:
min = 1
avg = 5.019693
max = 14
Number of instruments = 114 Wald chi2(10) = 458.13
Prob > chi2 = 0.0000
Two-step results
------------------------------------------------------------------------------
mdr | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
mdr |
L1. | .3819695 .0438487 8.71 0.000 .2960277 .4679113
|
lagebit_ta | .035684 .0136487 2.61 0.009 .008933 .062435
lagmb | .0147128 .0020845 7.06 0.000 .0106272 .0187983
lagdep_ta | .0648811 .091165 0.71 0.477 -.1137989 .2435612
laglnta | .030107 .0061345 4.91 0.000 .0180836 .0421304
lagfa_ta | .0150317 .0222106 0.68 0.499 -.0285004 .0585637
lagrd_dum | -.0178784 .0101335 -1.76 0.078 -.0377397 .0019829
lagrd_ta | .001471 .0313577 0.05 0.963 -.059989 .0629309
lagindmedian | .0919917 .0344887 2.67 0.008 .0243951 .1595883
lagrated | -.0066174 .0073056 -0.91 0.365 -.0209362 .0077014
------------------------------------------------------------------------------
Warning: gmm two-step standard errors are biased; robust standard
errors are recommended.
Instruments for differenced equation
GMM-type: L(2/.).mdr
Standard: D.lagebit_ta D.lagmb D.lagdep_ta D.laglnta D.lagfa_ta
D.lagrd_dum D.lagrd_ta D.lagindmedian D.lagrated
. est clear
. log close
name: SN
log: \5iexample10_s.smcl
log type: smcl
closed on: 5 Jun 2020, 01:48:02
----------------------------------------------------------------------------------------------------