xtivreg (validity of instruments and xtivreg2)

[nazlikaramollaoglu]

Dear all,
I have a firm level data (at id level) and I would like to use instrument2 as an instrument for sp_city_10t.
My instrument is at city level.

I estimate the following regression
xtset id year
xtivreg `var' i.year i.il_kodu (sp_city_10t=instrument2) , fe vce(cluster id)

I am wondering

1. what other additional test I have to do to make sure that my instrument is valid
2. what is the difference if I use xtivreg2 instead of xtivreg?

Thank you

[Jeff Wooldridge]

I meant to reply days ago. With only one IV you can’t use overid tests. You could use lagged or lead values of the IVs and add to your list. Then use the usual overid statistic.

[nazlikaramollaoglu]

I meant to reply days ago. With only one IV you can’t use overid tests. You could use lagged or lead values of the IVs and add to your list. Then use the usual overid statistic.

I have managed to install the xtivreg2 command and here is what I get when I ran the regression:

I am looking to the impact of immigration on firm level outcomes during the period 2006-2015.
I have instrumented the city level immigration density (im_den_city) with an instrument called instrument_p. Immigration starts at year 2013. before that year both immigration and instruments has a value of 0.

I have the following warnings:

1) How can I proceed to fix this problem?
Warning: estimated covariance matrix of moment conditions not of full rank.
overidentification statistic not reported, and standard errors and
model tests should be interpreted with caution.
Possible causes:
number of clusters insufficient to calculate robust covariance matrix
singleton dummy variable (dummy with one 1 and N-1 0s or vice versa)
partial option may address problem.

2) With regard to instrument_p it says that it has been excluded. I am not sure what included and excluded instruments at the end of my output means.



"Excluded instruments: instrument_p"

Thank you very much,

Nazlı

. foreach s of varlist man {
2. foreach x of varlist multi {
3. foreach z of varlist im_den_city {
4. foreach t of varlist reldsales_c {
5. local components "1m1 "
6. foreach y of local components {
7.
.
. xi:xtivreg2 `t' (`z'=instrument_p) i.year i.il_kodu if drop10_ncidef`y'!=1 & `x'==1 & `s'==1 , fe cluster (id)
8.
. }
9. }
10. }
11. }
12. }
i.year _Iyear_2006-2015 (naturally coded; _Iyear_2006 omitted)
i.il_kodu _Iil_kodu_1-81 (naturally coded; _Iil_kodu_1 omitted)
Warning - singleton groups detected. 19126 observation(s) not used.
Warning - collinearities detected
Vars dropped: _Iil_kodu_24 _Iil_kodu_30 _Iil_kodu_40 _Iil_kodu_56
_Iil_kodu_62 _Iil_kodu_75

FIXED EFFECTS ESTIMATION
------------------------
Number of groups = 46807 Obs per group: min = 2
avg = 5.1
max = 10
Warning - collinearities detected
Vars dropped: _Iil_kodu_24 _Iil_kodu_30 _Iil_kodu_40 _Iil_kodu_56 _Iil_kodu_62
_Iil_kodu_75

IV (2SLS) estimation
--------------------

Estimates efficient for homoskedasticity only
Statistics robust to heteroskedasticity and clustering on id

Number of clusters (id) = 46807 Number of obs = 237399
F( 84, 46806) = 99896.22
Prob > F = 0.0000
Total (centered) SS = 5069205.746 Centered R2 = 0.1053
Total (uncentered) SS = 5069205.746 Uncentered R2 = 0.1053
Residual SS = 4535499.256 Root MSE = 4.878

------------------------------------------------------------------------------
| Robust
reldsales_c | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
im_den_city | -.5762036 .0989316 -5.82 0.000 -.770106 -.3823012
_Iyear_2007 | .0976648 .0340531 2.87 0.004 .0309219 .1644076
_Iyear_2008 | -.0009134 .0451861 -0.02 0.984 -.0894765 .0876498
_Iyear_2009 | .096658 .0568388 1.70 0.089 -.014744 .20806
_Iyear_2010 | -.4168823 .0621761 -6.70 0.000 -.5387452 -.2950194
_Iyear_2011 | -.7950301 .0646489 -12.30 0.000 -.9217397 -.6683205
_Iyear_2012 | -1.020196 .0674435 -15.13 0.000 -1.152383 -.8880092
_Iyear_2013 | -.9302921 .0733194 -12.69 0.000 -1.073996 -.7865887
_Iyear_2014 | -.0683102 .1675632 -0.41 0.684 -.396728 .2601076
_Iyear_2015 | .3587017 .2340339 1.53 0.125 -.0999963 .8173998
_Iil_kodu_2 | 29.44216 15.43727 1.91 0.056 -.8143299 59.69864
_Iil_kodu_3 | 8.690306 5.278822 1.65 0.100 -1.655995 19.03661
_Iil_kodu_4 | 91.34342 2.706396 33.75 0.000 86.03898 96.64786
_Iil_kodu_5 | 48.80796 15.70783 3.11 0.002 18.02117 79.59475
_Iil_kodu_6 | -8.393589 3.064385 -2.74 0.006 -14.39967 -2.387506
_Iil_kodu_7 | -1.897948 4.526884 -0.42 0.675 -10.77048 6.974582
_Iil_kodu_8 | 84.20466 3.233215 26.04 0.000 77.86767 90.54164
_Iil_kodu_9 | 7.347658 6.79714 1.08 0.280 -5.974491 20.66981
_Iil_kodu_10 | -2.250266 3.957938 -0.57 0.570 -10.00768 5.50715
_Iil_kodu_11 | 30.87525 7.548087 4.09 0.000 16.08127 45.66923
_Iil_kodu_12 | 19.73949 9.05987 2.18 0.029 1.982474 37.49651
_Iil_kodu_13 | 86.39555 19.42912 4.45 0.000 48.31517 124.4759
_Iil_kodu_14 | 9.951167 7.597492 1.31 0.190 -4.939644 24.84198
_Iil_kodu_15 | 12.84179 7.422838 1.73 0.084 -1.706702 27.39029
_Iil_kodu_16 | -5.589553 2.914063 -1.92 0.055 -11.30101 .1219053
_Iil_kodu_17 | 26.20636 10.70143 2.45 0.014 5.231931 47.18078
_Iil_kodu_18 | 27.33962 9.662762 2.83 0.005 8.400955 46.27829
_Iil_kodu_19 | 20.59317 9.56091 2.15 0.031 1.854126 39.3322
_Iil_kodu_20 | 5.635091 8.006624 0.70 0.482 -10.0576 21.32779
_Iil_kodu_21 | 15.34765 10.44991 1.47 0.142 -5.133799 35.82911
_Iil_kodu_22 | 15.78745 8.178687 1.93 0.054 -.2424824 31.81738
_Iil_kodu_23 | -.6148941 4.92977 -0.12 0.901 -10.27707 9.047278
_Iil_kodu_24 | 0 (omitted)
_Iil_kodu_25 | 93.5898 2.914792 32.11 0.000 87.87691 99.30268
_Iil_kodu_26 | 1.787383 3.86693 0.46 0.644 -5.79166 9.366426
_Iil_kodu_27 | 9.358109 5.326706 1.76 0.079 -1.082043 19.79826
_Iil_kodu_28 | 15.13512 9.226111 1.64 0.101 -2.947721 33.21797
_Iil_kodu_29 | 50.06381 15.71621 3.19 0.001 19.2606 80.86701
_Iil_kodu_30 | 0 (omitted)
_Iil_kodu_31 | -.966562 4.928978 -0.20 0.845 -10.62718 8.694057
_Iil_kodu_32 | 15.90747 8.069256 1.97 0.049 .0920174 31.72292
_Iil_kodu_33 | 1.679033 3.576014 0.47 0.639 -5.329825 8.687891
_Iil_kodu_34 | -9.500203 2.707073 -3.51 0.000 -14.80597 -4.194438
_Iil_kodu_35 | -8.231145 3.22713 -2.55 0.011 -14.5562 -1.906087
_Iil_kodu_36 | 72.78962 3.22789 22.55 0.000 66.46308 79.11617
_Iil_kodu_37 | 28.97764 9.355766 3.10 0.002 10.64068 47.31461
_Iil_kodu_38 | 2.104457 3.428279 0.61 0.539 -4.614846 8.823761
_Iil_kodu_39 | 24.9803 6.165805 4.05 0.000 12.89554 37.06505
_Iil_kodu_40 | 0 (omitted)
_Iil_kodu_41 | -6.946914 2.740893 -2.53 0.011 -12.31897 -1.574863
_Iil_kodu_42 | -2.132555 3.659241 -0.58 0.560 -9.304536 5.039425
_Iil_kodu_43 | 18.73004 8.19139 2.29 0.022 2.675212 34.78487
_Iil_kodu_44 | 1.792785 4.194343 0.43 0.669 -6.427976 10.01355
_Iil_kodu_45 | -1.724836 3.584113 -0.48 0.630 -8.749568 5.299897
_Iil_kodu_46 | 8.197997 7.346484 1.12 0.264 -6.200847 22.59684
_Iil_kodu_47 | 55.57709 17.31028 3.21 0.001 21.64957 89.50461
_Iil_kodu_48 | 12.09063 5.790977 2.09 0.037 .7405226 23.44074
_Iil_kodu_49 | 30.12639 20.85611 1.44 0.149 -10.75083 71.00362
_Iil_kodu_50 | 34.41206 12.18093 2.83 0.005 10.53786 58.28625
_Iil_kodu_51 | 28.72148 11.185 2.57 0.010 6.799287 50.64367
_Iil_kodu_52 | 2.931459 5.135767 0.57 0.568 -7.134459 12.99738
_Iil_kodu_53 | -1.18564 8.071651 -0.15 0.883 -17.00579 14.6345
_Iil_kodu_54 | 6.575063 4.137674 1.59 0.112 -1.534629 14.68475
_Iil_kodu_55 | 10.7979 6.190758 1.74 0.081 -1.335759 22.93157
_Iil_kodu_56 | 0 (omitted)
_Iil_kodu_57 | 32.21007 11.13716 2.89 0.004 10.38164 54.03851
_Iil_kodu_58 | 9.569923 4.094442 2.34 0.019 1.544965 17.59488
_Iil_kodu_59 | -2.217016 2.859863 -0.78 0.438 -7.822245 3.388212
_Iil_kodu_60 | 13.01561 10.62499 1.23 0.221 -7.808982 33.84021
_Iil_kodu_61 | 24.55625 9.033255 2.72 0.007 6.851392 42.2611
_Iil_kodu_62 | 0 (omitted)
_Iil_kodu_63 | 28.65346 9.808822 2.92 0.003 9.428518 47.87839
_Iil_kodu_64 | 13.82003 6.043842 2.29 0.022 1.974321 25.66575
_Iil_kodu_65 | 48.83981 19.42984 2.51 0.012 10.75803 86.92159
_Iil_kodu_66 | 38.0031 13.58574 2.80 0.005 11.37553 64.63067
_Iil_kodu_67 | 13.1867 8.347137 1.58 0.114 -3.173388 29.54679
_Iil_kodu_68 | 38.12765 15.35253 2.48 0.013 8.03724 68.21805
_Iil_kodu_69 | 89.88754 2.708209 33.19 0.000 84.57955 95.19553
_Iil_kodu_70 | 5.99614 3.958578 1.51 0.130 -1.76253 13.75481
_Iil_kodu_71 | 35.56 11.5334 3.08 0.002 12.95496 58.16504
_Iil_kodu_72 | 9.60446 6.049997 1.59 0.112 -2.253316 21.46224
_Iil_kodu_73 | 82.84697 13.96029 5.93 0.000 55.4853 110.2086
_Iil_kodu_74 | 14.36211 16.82778 0.85 0.393 -18.61973 47.34395
_Iil_kodu_75 | 0 (omitted)
_Iil_kodu_76 | 103.5301 38.2773 2.70 0.007 28.50793 178.5522
_Iil_kodu_77 | 21.50701 7.554147 2.85 0.004 6.701157 36.31287
_Iil_kodu_78 | -5.289692 3.28385 -1.61 0.107 -11.72592 1.146535
_Iil_kodu_79 | 120.091 10.94817 10.97 0.000 98.63298 141.549
_Iil_kodu_80 | 22.03528 5.565299 3.96 0.000 11.1275 32.94307
_Iil_kodu_81 | 13.29335 4.761517 2.79 0.005 3.960952 22.62576
------------------------------------------------------------------------------
Underidentification test (Kleibergen-Paap rk LM statistic): 24.381
Chi-sq(1) P-val = 0.0000
------------------------------------------------------------------------------
Weak identification test (Cragg-Donald Wald F statistic): 4.3e+04
(Kleibergen-Paap rk Wald F statistic): 187.695
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.
------------------------------------------------------------------------------
Warning: estimated covariance matrix of moment conditions not of full rank.
overidentification statistic not reported, and standard errors and
model tests should be interpreted with caution.
Possible causes:
number of clusters insufficient to calculate robust covariance matrix
singleton dummy variable (dummy with one 1 and N-1 0s or vice versa)
partial option may address problem.
------------------------------------------------------------------------------
Instrumented: im_den_city
Included instruments: _Iyear_2007 _Iyear_2008 _Iyear_2009 _Iyear_2010
_Iyear_2011 _Iyear_2012 _Iyear_2013 _Iyear_2014
_Iyear_2015 _Iil_kodu_2 _Iil_kodu_3 _Iil_kodu_4
_Iil_kodu_5 _Iil_kodu_6 _Iil_kodu_7 _Iil_kodu_8
_Iil_kodu_9 _Iil_kodu_10 _Iil_kodu_11 _Iil_kodu_12
_Iil_kodu_13 _Iil_kodu_14 _Iil_kodu_15 _Iil_kodu_16
_Iil_kodu_17 _Iil_kodu_18 _Iil_kodu_19 _Iil_kodu_20
_Iil_kodu_21 _Iil_kodu_22 _Iil_kodu_23 _Iil_kodu_25
_Iil_kodu_26 _Iil_kodu_27 _Iil_kodu_28 _Iil_kodu_29
_Iil_kodu_31 _Iil_kodu_32 _Iil_kodu_33 _Iil_kodu_34
_Iil_kodu_35 _Iil_kodu_36 _Iil_kodu_37 _Iil_kodu_38
_Iil_kodu_39 _Iil_kodu_41 _Iil_kodu_42 _Iil_kodu_43
_Iil_kodu_44 _Iil_kodu_45 _Iil_kodu_46 _Iil_kodu_47
_Iil_kodu_48 _Iil_kodu_49 _Iil_kodu_50 _Iil_kodu_51
_Iil_kodu_52 _Iil_kodu_53 _Iil_kodu_54 _Iil_kodu_55
_Iil_kodu_57 _Iil_kodu_58 _Iil_kodu_59 _Iil_kodu_60
_Iil_kodu_61 _Iil_kodu_63 _Iil_kodu_64 _Iil_kodu_65
_Iil_kodu_66 _Iil_kodu_67 _Iil_kodu_68 _Iil_kodu_69
_Iil_kodu_70 _Iil_kodu_71 _Iil_kodu_72 _Iil_kodu_73
_Iil_kodu_74 _Iil_kodu_76 _Iil_kodu_77 _Iil_kodu_78
_Iil_kodu_79 _Iil_kodu_80 _Iil_kodu_81
Excluded instruments: instrument_p
Dropped collinear: _Iil_kodu_24 _Iil_kodu_30 _Iil_kodu_40 _Iil_kodu_56
_Iil_kodu_62 _Iil_kodu_75
------------------------------------------------------------------------------