Dear Statalisters,
I apply 2SLS to deal with/ address the heterogeneity concerns between competition (the Lerner index) and financial performance of our banks. For stage two of the 2SLS model we need to use the fitted values of Lerner index. How to generate these fitted values (that is, what command to use).
Below you can find my approach to run first and second stage:
I am using OLS regression for the first stage, where I put the lag for the instrumental variable lerner
Then I run 2SLS for the second stage by using ivreg2:
Will appreciate if you can tell me whether this is the right way to generate the so-called fitted value or I have to use
Thank you in advance for the help!
Petko Bachvarov
I apply 2SLS to deal with/ address the heterogeneity concerns between competition (the Lerner index) and financial performance of our banks. For stage two of the 2SLS model we need to use the fitted values of Lerner index. How to generate these fitted values (that is, what command to use).
Below you can find my approach to run first and second stage:
I am using OLS regression for the first stage, where I put the lag for the instrumental variable lerner
Code:
regress roa l.lerner es escovid19 lernercovid19 institution customerdeposit_growth loan_asset noninterestincome size corporateloansgrowthrate luqidasset equitytotalassets car gdp_growth inflation gcc_d d_iraq d_bahrain d_syrianarabrepublic d_palestinianterritories d_oman d_tunisia d_yemen d_saudiarabia d_jordan d_kuwait d_iran d_unitedarabemirates d_qatar d_lebanon d_egypt d_morocco d_libya d_algeria d_israel d_malta
note: d_qatar omitted because of collinearity
note: d_libya omitted because of collinearity
Source | SS df MS Number of obs = 2589
-------------+------------------------------ F( 34, 2554) = 5.95
Model | .44566372 34 .013107756 Prob > F = 0.0000
Residual | 5.62301178 2554 .002201649 R-squared = 0.0734
-------------+------------------------------ Adj R-squared = 0.0611
Total | 6.0686755 2588 .002344929 Root MSE = .04692
------------------------------------------------------------------------------------------
roa | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------------------+----------------------------------------------------------------
lerner |
L1. | -.0095362 .0036298 -2.63 0.009 -.0166539 -.0024185
|
es | .0060945 .0045187 1.35 0.178 -.0027662 .0149551
escovid19 | -.0224527 .0062567 -3.59 0.000 -.0347215 -.010184
lernercovid19 | .0399304 .0073166 5.46 0.000 .0255834 .0542775
institution | .0002905 .0014833 0.20 0.845 -.0026181 .003199
customerdeposit_growth | .0052759 .003515 1.50 0.133 -.0016167 .0121684
loan_asset | -.0042449 .0035936 -1.18 0.238 -.0112916 .0028017
noninterestincome | .0421363 .0060258 6.99 0.000 .0303204 .0539522
size | -.0001906 .0006648 -0.29 0.774 -.0014941 .001113
corporateloansgrowthrate | .0023772 .0037773 0.63 0.529 -.0050297 .009784
luqidasset | -.0046013 .0038684 -1.19 0.234 -.0121868 .0029843
equitytotalassets | .0020457 .0039702 0.52 0.606 -.0057394 .0098308
car | .0099154 .0049099 2.02 0.044 .0002876 .0195433
gdp_growth | -.0176409 .0195054 -0.90 0.366 -.055889 .0206072
inflation | .0049327 .0110903 0.44 0.657 -.0168141 .0266795
gcc_d | .0121645 .0103575 1.17 0.240 -.0081455 .0324745
d_iraq | .0017969 .0096179 0.19 0.852 -.0170628 .0206565
d_bahrain | -.0178029 .0069053 -2.58 0.010 -.0313434 -.0042623
d_syrianarabrepublic | .0463541 .0097503 4.75 0.000 .0272347 .0654734
d_palestinianterritories | .0055617 .0106894 0.52 0.603 -.015399 .0265224
d_oman | -.006375 .0072343 -0.88 0.378 -.0205607 .0078107
d_tunisia | .0048229 .0091034 0.53 0.596 -.0130279 .0226737
d_yemen | .0154679 .0135754 1.14 0.255 -.011152 .0420877
d_saudiarabia | .012205 .0075289 1.62 0.105 -.0025583 .0269683
d_jordan | .0080886 .0092123 0.88 0.380 -.0099758 .026153
d_kuwait | -.0069919 .0072173 -0.97 0.333 -.0211443 .0071605
d_iran | .0086982 .0104401 0.83 0.405 -.0117737 .0291702
d_unitedarabemirates | -.0031995 .0070636 -0.45 0.651 -.0170504 .0106514
d_qatar | 0 (omitted)
d_lebanon | .0042542 .0090399 0.47 0.638 -.0134721 .0219806
d_egypt | .0109324 .0089617 1.22 0.223 -.0066406 .0285053
d_morocco | .0037943 .00959 0.40 0.692 -.0150106 .0225993
d_libya | 0 (omitted)
d_algeria | .0147692 .0097202 1.52 0.129 -.0042911 .0338295
d_israel | -.0024029 .0102098 -0.24 0.814 -.0224233 .0176175
d_malta | .0132458 .0099935 1.33 0.185 -.0063504 .032842
_cons | .0036968 .0107959 0.34 0.732 -.0174728 .0248665
------------------------------------------------------------------------------------------
Code:
ivreg2 roa es escovid19 lernercovid19 institution customerdeposit_growth loan_asset noninterestincome size corporateloansgrowthrate luqidasset equitytotalassets car gdp_growth inflation (lerner=l.lerner) gcc_d d_iraq d_bahrain d_syrianarabrepublic d_palestinianterritories d_oman d_tunisia d_yemen d_saudiarabia d_jordan d_kuwait d_iran d_unitedarabemirates d_qatar d_lebanon d_egypt d_morocco d_libya d_algeria d_israel d_malta, endog(lerner)
Warning: time variable banks1 has 767 gap(s) in relevant range
Warning - collinearities detected
Vars dropped: d_qatar d_malta
IV (2SLS) estimation
--------------------
Estimates efficient for homoskedasticity only
Statistics consistent for homoskedasticity only
Number of obs = 2589
F( 34, 2554) = 2.45
Prob > F = 0.0000
Total (centered) SS = 6.068675495 Centered R2 = -1.2526
Total (uncentered) SS = 6.511511255 Uncentered R2 = -1.0994
Residual SS = 13.67057107 Root MSE = .07267
------------------------------------------------------------------------------------------
roa | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------------------+----------------------------------------------------------------
lerner | -.6281187 .370259 -1.70 0.090 -1.353813 .0975757
es | .0244415 .0129454 1.89 0.059 -.000931 .049814
escovid19 | -.072539 .0311774 -2.33 0.020 -.1336456 -.0114324
lernercovid19 | .111369 .0433862 2.57 0.010 .0263337 .1964043
institution | .0183552 .0108774 1.69 0.092 -.0029641 .0396745
customerdeposit_growth | .0162899 .0085851 1.90 0.058 -.0005367 .0331164
loan_asset | .0016047 .0066657 0.24 0.810 -.0114598 .0146693
noninterestincome | .0451403 .0094402 4.78 0.000 .0266379 .0636427
size | .0024808 .0018971 1.31 0.191 -.0012375 .0061992
corporateloansgrowthrate | -.0019177 .0063753 -0.30 0.764 -.0144131 .0105777
luqidasset | .0012117 .0066995 0.18 0.856 -.0119191 .0143425
equitytotalassets | .0057079 .0065284 0.87 0.382 -.0070876 .0185034
car | .014246 .007846 1.82 0.069 -.0011318 .0296238
gdp_growth | .011363 .03522 0.32 0.747 -.0576669 .0803929
inflation | .0594354 .0372514 1.60 0.111 -.013576 .1324468
gcc_d | .2015531 .1194758 1.69 0.092 -.0326152 .4357213
d_iraq | .4535345 .2741308 1.65 0.098 -.083752 .990821
d_bahrain | .2642198 .1674414 1.58 0.115 -.0639593 .592399
d_syrianarabrepublic | .4578509 .2506158 1.83 0.068 -.0333471 .9490489
d_palestinianterritories | .2953741 .1792197 1.65 0.099 -.05589 .6466382
d_oman | .0640446 .0428563 1.49 0.135 -.0199522 .1480413
d_tunisia | .3094563 .1871647 1.65 0.098 -.0573797 .6762924
d_yemen | .1987349 .1158293 1.72 0.086 -.0282864 .4257562
d_saudiarabia | -.132307 .08673 -1.53 0.127 -.3022947 .0376808
d_jordan | .2289398 .1379655 1.66 0.097 -.0414677 .4993473
d_kuwait | .2471417 .1505531 1.64 0.101 -.0479369 .5422204
d_iran | .0848352 .0536981 1.58 0.114 -.0204112 .1900816
d_unitedarabemirates | .0146991 .0151549 0.97 0.332 -.015004 .0444022
d_qatar | 0 (omitted)
d_lebanon | .5465863 .3271319 1.67 0.095 -.0945803 1.187753
d_egypt | .4988974 .2950886 1.69 0.091 -.0794656 1.07726
d_morocco | .3185791 .1928848 1.65 0.099 -.0594682 .6966264
d_libya | -.0047826 .0163956 -0.29 0.771 -.0369174 .0273522
d_algeria | .2218783 .1297559 1.71 0.087 -.0324386 .4761953
d_israel | .4710253 .2863757 1.64 0.100 -.0902607 1.032311
d_malta | 0 (omitted)
_cons | -.0379773 .0317201 -1.20 0.231 -.1001476 .024193
------------------------------------------------------------------------------------------
Underidentification test (Anderson canon. corr. LM statistic): 5.202
Chi-sq(1) P-val = 0.0226
------------------------------------------------------------------------------
Weak identification test (Cragg-Donald Wald F statistic): 5.142
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.
------------------------------------------------------------------------------
Sargan statistic (overidentification test of all instruments): 0.000
(equation exactly identified)
-endog- option:
Endogeneity test of endogenous regressors: 7.442
Chi-sq(1) P-val = 0.0064
Regressors tested: lerner
------------------------------------------------------------------------------
Instrumented: lerner
Included instruments: es escovid19 lernercovid19 institution
customerdeposit_growth loan_asset noninterestincome size
corporateloansgrowthrate luqidasset equitytotalassets car
gdp_growth inflation gcc_d d_iraq d_bahrain
d_syrianarabrepublic d_palestinianterritories d_oman
d_tunisia d_yemen d_saudiarabia d_jordan d_kuwait d_iran
d_unitedarabemirates d_lebanon d_egypt d_morocco d_libya
d_algeria d_israel
Excluded instruments: L.lerner
Dropped collinear: d_qatar d_malta
------------------------------------------------------------------------------
.
Code:
ivreg y (x1 = z1 z2) x2, first
Petko Bachvarov
Comment