Verbeek 5ed. Chapter 5 - Endogeneity, IV and GMM
Examples
----------------------------------------------------------------------------------------------------
name: SN
log: \5iexample5_s.smcl
log type: smcl
opened on: 9 Jun 2020, 23:38:34
. **********************************************
. * Solomon Negash - Examples
. * Verbeek(2017). A Giude To Modern Econometrics. 5ed.
. * STATA Program, version 16.1.
. * Chapter 5 - Endogeneity, Instrumental Variables and GMM
. ******************** **** *********************
. * Table 5.1 Wage equation estimated by OLS
. u "Data/schooling.dta", clear
. g exp76sq = exp76^2
. reg lwage76 ed76 exp76 exp76sq black smsa76 south76
Source | SS df MS Number of obs = 3,010
-------------+---------------------------------- F(6, 3003) = 204.93
Model | 172.16563 6 28.6942716 Prob > F = 0.0000
Residual | 420.476016 3,003 .140018653 R-squared = 0.2905
-------------+---------------------------------- Adj R-squared = 0.2891
Total | 592.641646 3,009 .196956346 Root MSE = .37419
------------------------------------------------------------------------------
lwage76 | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
ed76 | .074009 .0035054 21.11 0.000 .0671357 .0808823
exp76 | .0835958 .0066478 12.57 0.000 .0705612 .0966305
exp76sq | -.0022409 .0003178 -7.05 0.000 -.0028641 -.0016177
black | -.1896315 .0176266 -10.76 0.000 -.2241929 -.1550702
smsa76 | .161423 .0155733 10.37 0.000 .1308876 .1919583
south76 | -.1248615 .0151182 -8.26 0.000 -.1545046 -.0952184
_cons | 4.733664 .0676026 70.02 0.000 4.601112 4.866216
------------------------------------------------------------------------------
. *Table 5.2 Reduced form for schooling, estimated by OLS
. g age76sq=age76^2
. reg ed76 age76 age76sq black smsa76 south76 nearc4
Source | SS df MS Number of obs = 3,010
-------------+---------------------------------- F(6, 3003) = 67.29
Model | 2555.48762 6 425.914603 Prob > F = 0.0000
Residual | 19006.5924 3,003 6.32920161 R-squared = 0.1185
-------------+---------------------------------- Adj R-squared = 0.1168
Total | 21562.0801 3,009 7.16586243 Root MSE = 2.5158
------------------------------------------------------------------------------
ed76 | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
age76 | 1.061441 .3013985 3.52 0.000 .4704727 1.65241
age76sq | -.0187598 .0052314 -3.59 0.000 -.0290173 -.0085024
black | -1.468367 .1154434 -12.72 0.000 -1.694723 -1.242011
smsa76 | .8354027 .1092524 7.65 0.000 .6211856 1.04962
south76 | -.4596997 .1024337 -4.49 0.000 -.6605469 -.2588524
nearc4 | .347105 .1069972 3.24 0.001 .1373098 .5569002
_cons | -1.869524 4.298357 -0.43 0.664 -10.29755 6.558497
------------------------------------------------------------------------------
. * Table 5.3 Wage equation estimated by IV
. ivreg lwage76 (ed76=age76 age76sq nearc4) exp76 exp76sq black smsa76 south76
Instrumental variables (2SLS) regression
Source | SS df MS Number of obs = 3,010
-------------+---------------------------------- F(6, 3003) = 204.93
Model | 172.16563 6 28.6942716 Prob > F = 0.0000
Residual | 420.476016 3,003 .140018653 R-squared = 0.2905
-------------+---------------------------------- Adj R-squared = 0.2891
Total | 592.641646 3,009 .196956346 Root MSE = .37419
------------------------------------------------------------------------------
lwage76 | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
ed76 | .074009 .0035054 21.11 0.000 .0671357 .0808823
exp76 | .0835958 .0066478 12.57 0.000 .0705612 .0966305
exp76sq | -.0022409 .0003178 -7.05 0.000 -.0028641 -.0016177
black | -.1896315 .0176266 -10.76 0.000 -.2241929 -.1550702
smsa76 | .161423 .0155733 10.37 0.000 .1308876 .1919583
south76 | -.1248615 .0151182 -8.26 0.000 -.1545046 -.0952184
_cons | 4.733664 .0676026 70.02 0.000 4.601112 4.866216
------------------------------------------------------------------------------
Instrumented: ed76
Instruments: exp76 exp76sq black smsa76 south76 age76 age76sq nearc4
------------------------------------------------------------------------------
. * Table 5.4 OLS results explaining GDP per capita
. u "Data/institutions.dta", clear
. reg loggdp qi latitude
Source | SS df MS Number of obs = 64
-------------+---------------------------------- F(2, 61) = 41.18
Model | 39.3997135 2 19.6998568 Prob > F = 0.0000
Residual | 29.182005 61 .478393524 R-squared = 0.5745
-------------+---------------------------------- Adj R-squared = 0.5605
Total | 68.5817185 63 1.08859871 Root MSE = .69166
------------------------------------------------------------------------------
loggdp | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
qi | .4678871 .0641642 7.29 0.000 .3395827 .5961914
latitude | 1.576884 .71031 2.22 0.030 .1565313 2.997237
_cons | 4.728082 .3973208 11.90 0.000 3.93359 5.522574
------------------------------------------------------------------------------
. reg loggdp qi latitude africa asia malfal94
Source | SS df MS Number of obs = 62
-------------+---------------------------------- F(5, 56) = 31.90
Model | 48.1941941 5 9.63883882 Prob > F = 0.0000
Residual | 16.9204974 56 .30215174 R-squared = 0.7401
-------------+---------------------------------- Adj R-squared = 0.7169
Total | 65.1146915 61 1.06745396 Root MSE = .54968
------------------------------------------------------------------------------
loggdp | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
qi | .363567 .0562582 6.46 0.000 .2508684 .4762656
latitude | .2339876 .6253381 0.37 0.710 -1.018715 1.48669
africa | -.4137759 .2264297 -1.83 0.073 -.8673691 .0398174
asia | -.4569281 .2205636 -2.07 0.043 -.8987701 -.0150862
malfal94 | -.7876686 .2775151 -2.84 0.006 -1.343598 -.2317391
_cons | 6.177701 .4035055 15.31 0.000 5.369382 6.98602
------------------------------------------------------------------------------
. * Table 5.5 OLS results reduced form (QI explained from exogenous variables
. reg qi logem4 latitude
Source | SS df MS Number of obs = 64
-------------+---------------------------------- F(2, 61) = 12.82
Model | 40.2217245 2 20.1108623 Prob > F = 0.0000
Residual | 95.6645164 61 1.56827076 R-squared = 0.2960
-------------+---------------------------------- Adj R-squared = 0.2729
Total | 135.886241 63 2.15692446 Root MSE = 1.2523
------------------------------------------------------------------------------
qi | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
logem4 | -.5102681 .1410186 -3.62 0.001 -.7922521 -.228284
latitude | 2.001775 1.337176 1.50 0.140 -.6720747 4.675624
_cons | 8.529432 .8123128 10.50 0.000 6.905113 10.15375
------------------------------------------------------------------------------
. reg qi logem4 euro1900 latitude
Source | SS df MS Number of obs = 63
-------------+---------------------------------- F(3, 59) = 11.42
Model | 49.7402149 3 16.5800716 Prob > F = 0.0000
Residual | 85.6315451 59 1.45138212 R-squared = 0.3674
-------------+---------------------------------- Adj R-squared = 0.3353
Total | 135.37176 62 2.18341548 Root MSE = 1.2047
------------------------------------------------------------------------------
qi | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
logem4 | -.3684413 .1485276 -2.48 0.016 -.6656444 -.0712382
euro1900 | .0211563 .0082455 2.57 0.013 .0046571 .0376554
latitude | .2004581 1.495034 0.13 0.894 -2.791099 3.192015
_cons | 7.853131 .8308378 9.45 0.000 6.190628 9.515633
------------------------------------------------------------------------------
. reg qi logem4 latitude africa asia malfal94
Source | SS df MS Number of obs = 62
-------------+---------------------------------- F(5, 56) = 5.33
Model | 43.3212268 5 8.66424536 Prob > F = 0.0004
Residual | 91.0431828 56 1.62577112 R-squared = 0.3224
-------------+---------------------------------- Adj R-squared = 0.2619
Total | 134.36441 61 2.20269524 Root MSE = 1.2751
------------------------------------------------------------------------------
qi | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
logem4 | -.3283306 .199037 -1.65 0.105 -.7270497 .0703885
latitude | 1.887605 1.456952 1.30 0.200 -1.031022 4.806231
africa | .1351493 .5271787 0.26 0.799 -.9209166 1.191215
asia | .4874198 .5190193 0.94 0.352 -.5523008 1.52714
malfal94 | -.774151 .6952049 -1.11 0.270 -2.166814 .6185117
_cons | 7.871536 .9634927 8.17 0.000 5.941428 9.801643
------------------------------------------------------------------------------
. reg qi logem4 euro1900 latitude africa asia malfal94
Source | SS df MS Number of obs = 62
-------------+---------------------------------- F(6, 55) = 8.91
Model | 66.2240621 6 11.0373437 Prob > F = 0.0000
Residual | 68.1403475 55 1.23891541 R-squared = 0.4929
-------------+---------------------------------- Adj R-squared = 0.4375
Total | 134.36441 61 2.20269524 Root MSE = 1.1131
------------------------------------------------------------------------------
qi | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
logem4 | -.0313996 .1869719 -0.17 0.867 -.4060997 .3433005
euro1900 | .0437873 .0101841 4.30 0.000 .0233778 .0641967
latitude | -1.654378 1.51534 -1.09 0.280 -4.691188 1.382431
africa | 1.272259 .5307838 2.40 0.020 .2085447 2.335974
asia | 1.988919 .5720464 3.48 0.001 .8425125 3.135326
malfal94 | -1.241118 .6165232 -2.01 0.049 -2.476658 -.0055777
_cons | 5.861295 .9623004 6.09 0.000 3.932802 7.789788
------------------------------------------------------------------------------
. * Table 5.6 IV results explaining GDP per capita
. ivreg loggdp (qi=logem4) latitude
Instrumental variables (2SLS) regression
Source | SS df MS Number of obs = 64
-------------+---------------------------------- F(2, 61) = 17.01
Model | 7.02799121 2 3.5139956 Prob > F = 0.0000
Residual | 61.5537273 61 1.0090775 R-squared = 0.1025
-------------+---------------------------------- Adj R-squared = 0.0730
Total | 68.5817185 63 1.08859871 Root MSE = 1.0045
------------------------------------------------------------------------------
loggdp | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
qi | .995704 .2216816 4.49 0.000 .5524243 1.438984
latitude | -.6472071 1.335141 -0.48 0.630 -3.316986 2.022572
_cons | 1.691814 1.292985 1.31 0.196 -.8936697 4.277297
------------------------------------------------------------------------------
Instrumented: qi
Instruments: latitude logem4
------------------------------------------------------------------------------
. ivreg loggdp (qi=logem4 euro1900) latitude
Instrumental variables (2SLS) regression
Source | SS df MS Number of obs = 63
-------------+---------------------------------- F(2, 60) = 21.67
Model | 11.6179812 2 5.80899058 Prob > F = 0.0000
Residual | 55.0714197 60 .917856995 R-squared = 0.1742
-------------+---------------------------------- Adj R-squared = 0.1467
Total | 66.6894008 62 1.0756355 Root MSE = .95805
------------------------------------------------------------------------------
loggdp | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
qi | .9458258 .1733918 5.45 0.000 .5989906 1.292661
latitude | -.5971357 1.186311 -0.50 0.617 -2.970111 1.775839
_cons | 1.994732 1.017551 1.96 0.055 -.040674 4.030137
------------------------------------------------------------------------------
Instrumented: qi
Instruments: latitude logem4 euro1900
------------------------------------------------------------------------------
. ivreg loggdp (qi=logem4) latitude africa asia malfal94
Instrumental variables (2SLS) regression
Source | SS df MS Number of obs = 62
-------------+---------------------------------- F(5, 56) = 10.03
Model | 21.4528204 5 4.29056408 Prob > F = 0.0000
Residual | 43.6618711 56 .77967627 R-squared = 0.3295
-------------+---------------------------------- Adj R-squared = 0.2696
Total | 65.1146915 61 1.06745396 Root MSE = .88299
------------------------------------------------------------------------------
loggdp | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
qi | .8928218 .4198073 2.13 0.038 .0518467 1.733797
latitude | -1.069974 1.424529 -0.75 0.456 -3.923649 1.783701
africa | -.4452049 .3645429 -1.22 0.227 -1.175472 .2850623
asia | -.824812 .4546847 -1.81 0.075 -1.735655 .0860309
malfal94 | -.1057658 .6911819 -0.15 0.879 -1.49037 1.278838
_cons | 2.771564 2.716873 1.02 0.312 -2.670986 8.214114
------------------------------------------------------------------------------
Instrumented: qi
Instruments: latitude africa asia malfal94 logem4
------------------------------------------------------------------------------
. ivreg loggdp (qi=logem4 euro1900) latitude africa asia malfal94
Instrumental variables (2SLS) regression
Source | SS df MS Number of obs = 62
-------------+---------------------------------- F(5, 56) = 24.32
Model | 44.9497008 5 8.98994017 Prob > F = 0.0000
Residual | 20.1649907 56 .36008912 R-squared = 0.6903
-------------+---------------------------------- Adj R-squared = 0.6627
Total | 65.1146915 61 1.06745396 Root MSE = .60007
------------------------------------------------------------------------------
loggdp | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
qi | .5479184 .1147917 4.77 0.000 .317963 .7778737
latitude | -.2202115 .723272 -0.30 0.762 -1.669099 1.228676
africa | -.4247233 .2472542 -1.72 0.091 -.920033 .0705864
asia | -.5850704 .2500416 -2.34 0.023 -1.085964 -.0841767
malfal94 | -.5501465 .3277119 -1.68 0.099 -1.206632 .1063394
_cons | 4.991267 .7639299 6.53 0.000 3.460931 6.521602
------------------------------------------------------------------------------
Instrumented: qi
Instruments: latitude africa asia malfal94 logem4 euro1900
------------------------------------------------------------------------------
. * Table 5.7 GMM estimation results consumption-based asset pricing model
. log close
name: SN
log: \5iexample5_s.smcl
log type: smcl
closed on: 9 Jun 2020, 23:38:34
----------------------------------------------------------------------------------------------------