IV , 2SLS estimation problem

[Katerina Petrou]

Dear all,

Hello! I am using panel data and I did an Instrumental Variable estimation using the -xtivreg- command.
However, when I use the command -estat endog- to perform the test for endogeneity and the command -estat firststage- , Stata says that my commands are invalid but I am sure they are correct. Is there a way I can overcome this issue?
Thank you in advance.


Kind Regards,
Katerina
( Stata/SE 16.0)

[Andrew Musau]

However, when I use the command -estat endog- to perform the test for endogeneity and the command -estat firststage- , Stata says that my commands are invalid but I am sure they are correct.

These post estimation commands are available for ivregress and not xtivreg. If you are estimating a fixed effects IV model and the number of panels is not too large, you can include the fixed effects using dummies in ivregress.

[Katerina Petrou]

Andrew Musau Thank you so much for your answer... I didn't know that I am sorry. I will definitely do it again using ivregress.
Can you please explain to me how can I include the fixed effects using dummies?

[Andrew Musau]

webuse abdata
*PANEL IDENTIFIER= id
xtset id year
xtivreg n l2.n l(0/1).w l(0/2).(k ys) yr1981-yr1984 (l.n = l3.n), fe
*FIXED EFFECTS = i.id
ivregress 2sls n l2.n l(0/1).w l(0/2).(k ys) yr1981-yr1984 (l.n = l3.n) i.id, small

Res.:

. xtivreg n l2.n l(0/1).w l(0/2).(k ys) yr1981-yr1984 (l.n = l3.n), fe

Fixed-effects (within) IV regression            Number of obs     =        611
Group variable: id                              Number of groups  =        140

R-sq:                                           Obs per group:
     within  = 0.7288                                         min =          4
     between = 0.9913                                         avg =        4.4
     overall = 0.9856                                         max =          6

                                                Wald chi2(14)     =   54443.61
corr(u_i, Xb)  = 0.6854                         Prob > chi2       =     0.0000

------------------------------------------------------------------------------
           n |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
           n |
         L1. |   1.041278   .3025634     3.44   0.001      .448265    1.634292
         L2. |  -.3426033   .1621362    -2.11   0.035    -.6603845   -.0248221
             |
           w |
         --. |  -.6101005   .0708995    -8.61   0.000     -.749061   -.4711401
         L1. |   .5289202   .2060096     2.57   0.010     .1251488    .9326917
             |
           k |
         --. |   .3796942   .0497241     7.64   0.000     .2822368    .4771517
         L1. |  -.1949596   .1069424    -1.82   0.068    -.4045628    .0146436
         L2. |   .0020901   .0420389     0.05   0.960    -.0803046    .0844847
             |
          ys |
         --. |   .5587552   .1356942     4.12   0.000     .2927996    .8247109
         L1. |  -.9432021   .2605708    -3.62   0.000    -1.453912   -.4324927
         L2. |   .1832405   .2157935     0.85   0.396     -.239707     .606188
             |
      yr1981 |  -.0446299   .0175524    -2.54   0.011    -.0790319   -.0102279
      yr1982 |  -.0575767   .0221038    -2.60   0.009    -.1008994    -.014254
      yr1983 |  -.0647932   .0265031    -2.44   0.014    -.1167382   -.0128481
      yr1984 |  -.0589397   .0307644    -1.92   0.055    -.1192368    .0013573
       _cons |   1.618841   1.085379     1.49   0.136    -.5084637    3.746146
-------------+----------------------------------------------------------------
     sigma_u |  .17858284
     sigma_e |  .10448489
         rho |  .74498093   (fraction of variance due to u_i)
------------------------------------------------------------------------------
F  test that all u_i=0:     F(139,457) =     0.54         Prob > F    = 1.0000
------------------------------------------------------------------------------
Instrumented:   L.n
Instruments:    L2.n w L.w k L.k L2.k ys L.ys L2.ys yr1981 yr1982 yr1983
                yr1984 L3.n
------------------------------------------------------------------------------


. ivregress 2sls n l2.n l(0/1).w l(0/2).(k ys) yr1981-yr1984 (l.n = l3.n) i.id, small


Instrumental variables (2SLS) regression

      Source |       SS       df       MS         Number of obs   =        611
-------------+------------------------------      F(153,   457)   =     654.67
       Model |  1094.25433   153  7.15198907      Prob > F        =     0.0000
    Residual |  4.98911115   457  .010917092      R-squared       =     0.9955
-------------+------------------------------      Adj R-squared   =     0.9939
       Total |  1099.24344   610  1.80203842      Root MSE        =     .10448

------------------------------------------------------------------------------
           n |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
           n |
         L1. |   1.041278   .3025634     3.44   0.001     .4466903    1.635866
         L2. |  -.3426033   .1621362    -2.11   0.035    -.6612283   -.0239783
             |
           w |
         --. |  -.6101005   .0708995    -8.61   0.000      -.74943   -.4707711
         L1. |   .5289202   .2060096     2.57   0.011     .1240766    .9337639
             |
           k |
         --. |   .3796942   .0497241     7.64   0.000      .281978    .4774105
         L1. |  -.1949596   .1069424    -1.82   0.069    -.4051194    .0152002
         L2. |   .0020901   .0420389     0.05   0.960    -.0805234    .0847035
             |
          ys |
         --. |   .5587552   .1356942     4.12   0.000     .2920934    .8254171
         L1. |  -.9432021   .2605708    -3.62   0.000    -1.455268   -.4311366
         L2. |   .1832405   .2157935     0.85   0.396    -.2408301    .6073111
             |
      yr1981 |  -.0446299   .0175524    -2.54   0.011    -.0791232   -.0101366
      yr1982 |  -.0575767   .0221038    -2.60   0.009    -.1010144   -.0141389
      yr1983 |  -.0647932   .0265031    -2.44   0.015    -.1168761   -.0127102
      yr1984 |  -.0589397   .0307644    -1.92   0.056    -.1193969    .0015174
             |
          id |
          2  |   .2517751   .1806031     1.39   0.164    -.1031405    .6066907
          3  |   .0742088   .1449075     0.51   0.609    -.2105589    .3589765
          4  |   .1305803   .1380529     0.95   0.345    -.1407169    .4018775
          5  |   .2172429   .1983619     1.10   0.274    -.1725716    .6070575
          6  |  -.1555884   .1218008    -1.28   0.202    -.3949475    .0837706
          7  |  -.2962532   .1296279    -2.29   0.023    -.5509938   -.0415126
          8  |  -.0871508   .1383186    -0.63   0.529    -.3589702    .1846686
          9  |  -.0067442   .1202581    -0.06   0.955    -.2430715    .2295832
         10  |  -.1124819   .0980315    -1.15   0.252    -.3051303    .0801666
         11  |  -.2698972   .1240247    -2.18   0.030    -.5136266   -.0261678
         12  |  -.1306666   .1559439    -0.84   0.403    -.4371227    .1757895
         13  |  -.0659484   .1142375    -0.58   0.564    -.2904443    .1585475
         14  |  -.0076793   .1323763    -0.06   0.954     -.267821    .2524623
         15  |  -.0111962   .1011815    -0.11   0.912    -.2100349    .1876425
         16  |  -.0872834   .1137412    -0.77   0.443     -.310804    .1362372
         17  |   .0252075   .1513951     0.17   0.868    -.2723094    .3227245
         18  |   .0474451   .1317335     0.36   0.719    -.2114335    .3063237
         19  |   .1131747   .1829626     0.62   0.537    -.2463776    .4727271
         20  |   -.189771   .1195843    -1.59   0.113    -.4247742    .0452322
         21  |  -.1305493   .1112797    -1.17   0.241    -.3492327    .0881341
         22  |  -.1091228   .1394027    -0.78   0.434    -.3830725    .1648269
         23  |   .0783809   .1578123     0.50   0.620    -.2317467    .3885086
         24  |  -.1745129   .1015761    -1.72   0.086     -.374127    .0251012
         25  |  -.0655963   .1928929    -0.34   0.734    -.4446634    .3134708
         26  |  -.0514218   .1305443    -0.39   0.694    -.3079633    .2051196
         27  |   .0531983   .1167084     0.46   0.649    -.1761532    .2825499
         28  |  -.1240193   .1327934    -0.93   0.351    -.3849808    .1369422
         29  |  -.0954574   .1609571    -0.59   0.553    -.4117653    .2208505
         30  |  -.1799296   .1283232    -1.40   0.162    -.4321063    .0722471
         31  |  -.1120499   .1172216    -0.96   0.340    -.3424101    .1183104
         32  |   .0694667   .1611438     0.43   0.667     -.247208    .3861413
         33  |   .0560435    .166368     0.34   0.736    -.2708977    .3829846
         34  |   .0018361   .1196153     0.02   0.988    -.2332282    .2369004
         35  |   .0577946   .1551659     0.37   0.710    -.2471326    .3627217
         36  |  -.4621035   .1629334    -2.84   0.005     -.782295    -.141912
         37  |   .1129081   .2028406     0.56   0.578    -.2857078     .511524
         38  |  -.0122633   .1247676    -0.10   0.922    -.2574527    .2329261
         39  |  -.1969665   .1141827    -1.73   0.085    -.4213547    .0274217
         40  |   .2244004   .1926922     1.16   0.245    -.1542722    .6030729
         41  |   .2039286   .1998806     1.02   0.308    -.1888706    .5967277
         42  |  -.2274981   .1022853    -2.22   0.027    -.4285059   -.0264903
         43  |   .1804178   .1357542     1.33   0.185    -.0863621    .4471976
         44  |   -.217986   .1248449    -1.75   0.081    -.4633273    .0273554
         45  |   .1387278   .1819521     0.76   0.446    -.2188387    .4962943
         46  |  -.1330853   .1013107    -1.31   0.190    -.3321779    .0660074
         47  |  -.1849615   .1036917    -1.78   0.075    -.3887332    .0188101
         48  |  -.0882266   .1088925    -0.81   0.418    -.3022187    .1257656
         49  |   -.119241   .0931294    -1.28   0.201    -.3022559    .0637739
         50  |   .1521085   .1958487     0.78   0.438    -.2327671    .5369841
         51  |   -.211424   .1319087    -1.60   0.110    -.4706468    .0477989
         52  |   .0760409   .1366742     0.56   0.578     -.192547    .3446288
         53  |   .0548175   .1212921     0.45   0.652    -.1835419    .2931769
         54  |   .0380241   .1340204     0.28   0.777    -.2253486    .3013967
         55  |   .2472499    .312324     0.79   0.429    -.3665193    .8610192
         56  |   -.076495   .1036926    -0.74   0.461    -.2802686    .1272785
         57  |   -.011134   .1129377    -0.10   0.922    -.2330755    .2108076
         58  |   .0009457   .1481525     0.01   0.995    -.2901989    .2920902
         59  |   .1179909   .1089414     1.08   0.279    -.0960974    .3320791
         60  |  -.1371243   .1153977    -1.19   0.235    -.3639003    .0896516
         61  |  -.0958465    .099463    -0.96   0.336    -.2913081    .0996152
         62  |    .146384   .1645385     0.89   0.374    -.1769619    .4697299
         63  |  -.3517004   .1430039    -2.46   0.014    -.6327271   -.0706738
         64  |   .0785623    .169338     0.46   0.643    -.2542153    .4113399
         65  |   .0883213   .1752025     0.50   0.614     -.255981    .4326237
         66  |  -.1588166   .1429613    -1.11   0.267    -.4397596    .1221264
         67  |   .3281458   .2203723     1.49   0.137     -.104923    .7612145
         68  |  -.2390496   .1534602    -1.56   0.120    -.5406248    .0625255
         69  |  -.2080735   .1132986    -1.84   0.067    -.4307245    .0145774
         70  |  -.1627025   .1043485    -1.56   0.120    -.3677648    .0423599
         71  |  -.1923052   .1025127    -1.88   0.061      -.39376    .0091495
         72  |   .0978864   .1998232     0.49   0.624    -.2947999    .4905727
         73  |  -.2119449   .1373973    -1.54   0.124    -.4819537    .0580639
         74  |   -.045933   .1017244    -0.45   0.652    -.2458387    .1539726
         75  |  -.1109454   .1024858    -1.08   0.280    -.3123473    .0904565
         76  |   .0346132   .1362435     0.25   0.800    -.2331283    .3023547
         77  |    .030702   .0904099     0.34   0.734    -.1469687    .2083726
         78  |   .1450491   .1713988     0.85   0.398    -.1917784    .4818766
         79  |  -.2445013    .119748    -2.04   0.042    -.4798262   -.0091763
         80  |  -.1153298   .0962545    -1.20   0.231     -.304486    .0738265
         81  |  -.1811491   .1210509    -1.50   0.135    -.4190345    .0567364
         82  |    .187199   .2518328     0.74   0.458    -.3076949    .6820929
         83  |  -.0328778   .1152103    -0.29   0.775    -.2592855    .1935299
         84  |   .0373874   .1222552     0.31   0.760    -.2028647    .2776396
         85  |   .0945172     .14233     0.66   0.507    -.1851852    .3742197
         86  |   .1603459   .2021749     0.79   0.428    -.2369617    .5576536
         87  |   .1695656   .1902668     0.89   0.373    -.2043407    .5434719
         88  |   .0537142   .1625898     0.33   0.741    -.2658021    .3732305
         89  |  -.4377228   .1791594    -2.44   0.015    -.7898012   -.0856444
         90  |  -.0972459   .1467031    -0.66   0.508    -.3855422    .1910503
         91  |   -.432329   .1709619    -2.53   0.012     -.768298   -.0963601
         92  |  -.2886952   .1839772    -1.57   0.117    -.6502413    .0728509
         93  |   .2189444    .207467     1.06   0.292    -.1887632    .6266521
         94  |  -.0731296   .1368757    -0.53   0.593    -.3421134    .1958541
         95  |  -.2033389   .1233175    -1.65   0.100    -.4456787    .0390008
         96  |   .2295656    .214939     1.07   0.286    -.1928257    .6519569
         97  |   .1539901   .1735061     0.89   0.375    -.1869787    .4949589
         98  |  -.1097122   .1501563    -0.73   0.465    -.4047945    .1853702
         99  |  -.2204389   .1188276    -1.86   0.064    -.4539552    .0130775
        100  |    .058274   .1111775     0.52   0.600    -.1602086    .2767565
        101  |  -.0139588   .0918845    -0.15   0.879    -.1945274    .1666098
        102  |  -.3098431    .134527    -2.30   0.022    -.5742115   -.0454748
        103  |  -.3786854   .1405813    -2.69   0.007    -.6549513   -.1024195
        104  |  -.3601789   .1595082    -2.26   0.024    -.6736394   -.0467184
        105  |  -.5829487   .2156984    -2.70   0.007    -1.006832   -.1590649
        106  |  -.1918637   .1731304    -1.11   0.268    -.5320941    .1483666
        107  |  -.2754172   .2114836    -1.30   0.193    -.6910182    .1401838
        108  |  -.5198031   .2396777    -2.17   0.031    -.9908102   -.0487959
        109  |  -.0582931   .1237696    -0.47   0.638    -.3015213     .184935
        110  |  -.0143063   .1073836    -0.13   0.894    -.2253332    .1967207
        111  |  -.0578274   .1152043    -0.50   0.616    -.2842234    .1685685
        112  |  -.0363783   .1164905    -0.31   0.755    -.2653017    .1925451
        113  |  -.1281844   .1265843    -1.01   0.312     -.376944    .1205751
        114  |  -.0482467   .1479883    -0.33   0.745    -.3390687    .2425752
        115  |   .0007824    .111764     0.01   0.994    -.2188527    .2204175
        116  |  -.0213621   .0985384    -0.22   0.828    -.2150066    .1722825
        117  |  -.0335467   .0986006    -0.34   0.734    -.2273134    .1602201
        118  |   .0995199   .0837558     1.19   0.235    -.0650743    .2641142
        119  |  -.1028155   .1214693    -0.85   0.398    -.3415232    .1358921
        120  |  -.0898806   .1128328    -0.80   0.426    -.3116161     .131855
        121  |  -.1221363   .1107666    -1.10   0.271    -.3398112    .0955387
        122  |  -.0695613   .1179351    -0.59   0.556    -.3013236     .162201
        123  |  -.2156587   .1184343    -1.82   0.069     -.448402    .0170846
        124  |  -.5070985   .1990999    -2.55   0.011    -.8983635   -.1158336
        125  |  -.2471386   .1402292    -1.76   0.079    -.5227126    .0284353
        126  |  -.3741152   .1782221    -2.10   0.036    -.7243516   -.0238789
        127  |  -.3558652   .1997056    -1.78   0.075    -.7483203    .0365899
        128  |  -.2319358     .12087    -1.92   0.056    -.4694657     .005594
        129  |  -.3654399   .2424295    -1.51   0.132    -.8418546    .1109749
        130  |  -.3891675   .2588038    -1.50   0.133    -.8977606    .1194257
        131  |  -.0717827     .10572    -0.68   0.497    -.2795403    .1359748
        132  |  -.1216998   .1172277    -1.04   0.300    -.3520719    .1086723
        133  |   .0041288   .1323812     0.03   0.975    -.2560226    .2642803
        134  |   .2516704   .1398313     1.80   0.073    -.0231218    .5264625
        135  |  -.0935282   .0845209    -1.11   0.269    -.2596261    .0725696
        136  |  -.2371663   .1395315    -1.70   0.090    -.5113692    .0370365
        137  |  -.2474227   .1188542    -2.08   0.038    -.4809911   -.0138542
        138  |  -.0144718   .1459152    -0.10   0.921    -.3012199    .2722762
        139  |  -.1841477   .0972079    -1.89   0.059    -.3751776    .0068821
        140  |  -.2335365   .1131343    -2.06   0.040    -.4558644   -.0112087
             |
       _cons |   1.702652   1.052304     1.62   0.106    -.3653019    3.770606
------------------------------------------------------------------------------
Instrumented:  L.n
Instruments:   L2.n w L.w k L.k L2.k ys L.ys L2.ys yr1981 yr1982 yr1983
               yr1984 2.id 3.id 4.id 5.id 6.id 7.id 8.id 9.id 10.id 11.id
               12.id 13.id 14.id 15.id 16.id 17.id 18.id 19.id 20.id 21.id
               22.id 23.id 24.id 25.id 26.id 27.id 28.id 29.id 30.id 31.id
               32.id 33.id 34.id 35.id 36.id 37.id 38.id 39.id 40.id 41.id
               42.id 43.id 44.id 45.id 46.id 47.id 48.id 49.id 50.id 51.id
               52.id 53.id 54.id 55.id 56.id 57.id 58.id 59.id 60.id 61.id
               62.id 63.id 64.id 65.id 66.id 67.id 68.id 69.id 70.id 71.id
               72.id 73.id 74.id 75.id 76.id 77.id 78.id 79.id 80.id 81.id
               82.id 83.id 84.id 85.id 86.id 87.id 88.id 89.id 90.id 91.id
               92.id 93.id 94.id 95.id 96.id 97.id 98.id 99.id 100.id 101.id
               102.id 103.id 104.id 105.id 106.id 107.id 108.id 109.id
               110.id 111.id 112.id 113.id 114.id 115.id 116.id 117.id
               118.id 119.id 120.id 121.id 122.id 123.id 124.id 125.id
               126.id 127.id 128.id 129.id 130.id 131.id 132.id 133.id
               134.id 135.id 136.id 137.id 138.id 139.id 140.id L3.n

.

[Katerina Petrou]

Andrew Musau Thank you so much for your answer but I am a bit confused as I am not an expert in Stata.
Before I used the command:

xtivreg lngdp (tariff=tariff_L1), fe

So, now, should I use this?

ivregress 2sls lngdp (tariff=tariff_L1) i.id, small

[Jeff Wooldridge]

Katerina: Unfortunately, Stata has not incorporated some simple tests for endogeneity for xtivregress. But it is easy to obtain a control function version of the test, and also make it robust to serial correlation and heteroskedasticity, as you should. To obtain the FEIV estimates and proper standard errors, with w endogenous:

xtivreg y (w = z1 ... zM) x1 ... xK i.year, vce(cluster id)

x1 ... xK are included exogenous variables, z1 ... zM are excluded instruments. Here is how to test the null that w is exogenous:

xtreg w z1 ... zM x1 ... xK i.year
predict vhat, e
xtreg y w vhat x1 ... xK i.year, vce(cluster id)

The robust t statistic on vhat is the test of the null that w is exogenous. This test is described in Chapter 11 of my 2010 MIT Press book. As a check that you did it properly, the second xtreg command should produce the same estimates as xtivreg (except for the coefficient on vhat, of course, since there is no comparable coefficient with xtivreg). If the panel is unbalanced you have to be careful to use only the complete cases in the two xtreg commands.

Andrew's idea probably works, too, but I'm hesitant to just add dummies and use traditional commands.

[Andrew Musau]

xtivreg lngdp (tariff=tariff_L1), fe

So, now, should I use this?
ivregress 2sls lngdp (tariff=tariff_L1) i.id, small

Yes, make sure you get the same estimates for the regression coefficients (see #4). You should implement Jeff Wooldridge's approach and only use this as a check, but I think the results will be consistent.

[Katerina Petrou]

Jeff Wooldridge Thank you so much for your answer, I really appreciate it.
I have tried everything you said and for vhat I got:
| Robust
lngdp | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
vhat | .0069672 .0056433 1.23 0.217 -.0040935 .0180279

Is this a good thing? I am not completely sure.

[Katerina Petrou]

Andrew Musau Thank you for your answer.
Unfortunately, I have tried both commands in

#5 and

the first gives me a negative coefficient and the second one gives me a positive one.
What I prefer is the negative one. Is there something I can do about it?
I like the idea using the postestimation tests after the ivregress but I get a positive coefficient.

[Andrew Musau]

Unfortunately, I have tried both commands in
#5 and
the first gives me a negative coefficient and the second one gives me a positive one.

Something is not right. You should get the same coefficient. Can you copy and paste all the commands that you entered (including the xtset) and the regression outputs. Post this between CODE delimiters

\(\text{[CODE ]}\)
STATA OUTPUT
\(\text{[/CODE ]}\)

[Katerina Petrou]

Andrew Musau , you can find them below.

 xtset ifscode year
       panel variable:  ifscode (strongly balanced)
        time variable:  year, 1960 to 2014
                delta:  1 unit

. gen lngdp = ln(ngdp_r)
(45 missing values generated)

. gen tariff_L1=tariff[_n-1]
(374 missing values generated)


. xtivreg lngdp (tariff=tariff_L1) ,fe

Fixed-effects (within) IV regression            Number of obs     =      1,303
Group variable: ifscode                         Number of groups  =         31

R-sq:                                           Obs per group:
     within  = 0.4551                                         min =         12
     between = 0.0795                                         avg =       42.0
     overall = 0.0029                                         max =         55

                                                Wald chi2(1)      =  356535.87
corr(u_i, Xb)  = -0.2300                        Prob > chi2       =     0.0000

------------------------------------------------------------------------------
       lngdp |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
      tariff |  -.0649125   .0020939   -31.00   0.000    -.0690166   -.0608084
       _cons |   6.633049   .0162812   407.40   0.000     6.601138     6.66496
-------------+----------------------------------------------------------------
     sigma_u |  2.6785784
     sigma_e |  .37814296
         rho |  .98045962   (fraction of variance due to u_i)
------------------------------------------------------------------------------
F  test that all u_i=0:     F(30,1271) =  1835.74         Prob > F    = 0.0000
------------------------------------------------------------------------------
Instrumented:   tariff
Instruments:    tariff_L1
------------------------------------------------------------------------------






. ivregress 2sls lngdp (tariff=tariff_L1) i.year, small


Instrumental variables (2SLS) regression

      Source |       SS       df       MS         Number of obs   =      1,303
-------------+------------------------------      F( 55,  1247)   =       1.33
       Model |  446.622692    55  8.12041259      Prob > F        =     0.0579
    Residual |  7691.93207  1247   6.1683497      R-squared       =     0.0549
-------------+------------------------------      Adj R-squared   =     0.0132
       Total |  8138.55476  1302  6.25081011      Root MSE        =     2.4836

------------------------------------------------------------------------------
       lngdp |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
      tariff |   .0788904    .012817     6.16   0.000     .0537452    .1040356
             |
        year |
       1961  |   .8159265   1.052261     0.78   0.438    -1.248471    2.880324
       1962  |   .9066259   1.052922     0.86   0.389    -1.159069     2.97232
       1963  |   1.022899   1.054032     0.97   0.332    -1.044974    3.090771
       1964  |   1.129749   1.055028     1.07   0.284    -.9400761    3.199574
       1965  |    1.17065   1.054509     1.11   0.267    -.8981581    3.239458
       1966  |   1.300966    1.05629     1.23   0.218    -.7713353    3.373268
       1967  |   1.350423   1.056441     1.28   0.201    -.7221758    3.423022
       1968  |   1.438245   1.057128     1.36   0.174    -.6357014    3.512192
       1969  |   1.584886   1.059023     1.50   0.135    -.4927783     3.66255
       1970  |    1.70358   1.060723     1.61   0.109     -.377418    3.784578
       1971  |   1.813983    1.06223     1.71   0.088    -.2699728    3.897938
       1972  |   1.839835   1.042215     1.77   0.078    -.2048535    3.884524
       1973  |   1.878615   1.041403     1.80   0.071    -.1644813     3.92171
       1974  |   1.969647   1.028279     1.92   0.056     -.047701    3.986994
       1975  |   1.902601   1.025975     1.85   0.064    -.1102258    3.915428
       1976  |   1.906318   1.024307     1.86   0.063     -.103237    3.915873
       1977  |   1.924006   1.015972     1.89   0.058    -.0691979    3.917209
       1978  |   2.051024    1.01865     2.01   0.044     .0525668     4.04948
       1979  |   2.163197   1.020961     2.12   0.034      .160207    4.166187
       1980  |   2.255895   1.023258     2.20   0.028     .2483981    4.263391
       1981  |   2.044908   1.016989     2.01   0.045     .0497101    4.040105
       1982  |   2.138942   1.012479     2.11   0.035     .1525929    4.125292
       1983  |   2.165623   1.003052     2.16   0.031     .1977681    4.133478
       1984  |   2.219384   1.003405     2.21   0.027     .2508366    4.187932
       1985  |   2.297833   1.005184     2.29   0.022     .3257952    4.269872
       1986  |   2.290426   1.003539     2.28   0.023     .3216143    4.259238
       1987  |   2.302778   1.002334     2.30   0.022     .3363314    4.269225
       1988  |   2.538044   1.009816     2.51   0.012     .5569183    4.519169
       1989  |   2.522015   .9836492     2.56   0.010      .592225    4.451805
       1990  |   2.572993   .9783864     2.63   0.009     .6535278    4.492458
       1991  |   2.602052   .9791122     2.66   0.008     .6811634    4.522941
       1992  |   2.642177   .9804672     2.69   0.007     .7186299    4.565725
       1993  |   2.680798   .9815611     2.73   0.006     .7551042    4.606491
       1994  |   2.718264   .9813699     2.77   0.006      .792946    4.643583
       1995  |   2.651862   .9768231     2.71   0.007     .7354635     4.56826
       1996  |   2.798248   .9816321     2.85   0.004      .872415    4.724081
       1997  |   2.881779   .9835265     2.93   0.003     .9522297    4.811328
       1998  |   2.939394   .9847516     2.98   0.003     1.007441    4.871347
       1999  |   3.006955    .985953     3.05   0.002     1.072646    4.941265
       2000  |   3.039028   .9852042     3.08   0.002     1.106187    4.971869
       2001  |   3.011187   .9830183     3.06   0.002     1.082634    4.939739
       2002  |   3.081831   .9851914     3.13   0.002     1.149015    5.014647
       2003  |   3.111047   .9854794     3.16   0.002     1.177667    5.044428
       2004  |   3.158497   .9858609     3.20   0.001     1.224368    5.092626
       2005  |   3.213008   .9867248     3.26   0.001     1.277184    5.148832
       2006  |   3.259604   .9869905     3.30   0.001     1.323259     5.19595
       2007  |   3.305041   .9871903     3.35   0.001     1.368304    5.241778
       2008  |    3.32016   .9876214     3.36   0.001     1.382577    5.257743
       2009  |   3.278594   .9875772     3.32   0.001     1.341098     5.21609
       2010  |   3.305102   .9910403     3.33   0.001     1.360812    5.249393
       2011  |   3.335885   .9916146     3.36   0.001     1.390468    5.281303
       2012  |   3.430579    .992699     3.46   0.001     1.483034    5.378123
       2013  |   3.425289   .9920791     3.45   0.001     1.478961    5.371618
       2014  |   4.583692   1.058543     4.33   0.000      2.50697    6.660414
             |
       _cons |   3.219413    .889112     3.62   0.000     1.475092    4.963733
------------------------------------------------------------------------------
Instrumented:  tariff
Instruments:   1961.year 1962.year 1963.year 1964.year 1965.year 1966.year
               1967.year 1968.year 1969.year 1970.year 1971.year 1972.year
               1973.year 1974.year 1975.year 1976.year 1977.year 1978.year
               1979.year 1980.year 1981.year 1982.year 1983.year 1984.year
               1985.year 1986.year 1987.year 1988.year 1989.year 1990.year
               1991.year 1992.year 1993.year 1994.year 1995.year 1996.year
               1997.year 1998.year 1999.year 2000.year 2001.year 2002.year
               2003.year 2004.year 2005.year 2006.year 2007.year 2008.year
               2009.year 2010.year 2011.year 2012.year 2013.year 2014.year
               tariff_L1

[Jeff Wooldridge]

Katerina: You have a couple of mistakes. First, you will want to put i.year in the xtivreg command. What xtivreg does is account for cross-sectional fixed effects. Second, you forgot i.id in the ivregress command. That is how you include the cross-sectional fixed effects. So, the way to do this, generically, is

xtset ifscode year
xtivreg y (w = z1 ... zM) x1 ... xK i.year, fe
ivregress 2sls y (w = z1 ... zM) x1 ... xK i.year i.ifscode

The standard errors might not be the same because of different conventions with degrees of freedom, but the estimates should be.

Your results from xtivreg are not believable until you include id.year and also cluster the standard errors. Even then, it seems odd you don't have more control variables.

And I just noticed that your T is bigger than your N. This is not the traditional panel data setup, and computing standard errors can be tricky. Is this a research project for a dissertation?

[Katerina Petrou]

Jeff Wooldridge Thank you very much for the code. Now I get the same coefficients!

Yes, T is bigger than N...I thought that this was ok but it seems it isn't...Should I proceed with the IV estimation?
and yes this a research project for an undergraduate dissertation.

[Jeff Wooldridge]

Using the estimator is fine. although there may be problems with unit roots. But clustering by ifscode when N = 31 is borderline.

[Katerina Petrou]

Jeff Wooldridge I will definitely have this in mind now. Thank you very much!