Announcement

Collapse
No announcement yet.
X
Collapse
  • Page of 1
  • Filter
  • Time
  • Show
Clear All
new posts
  • #1

    First stage difference xtivreg vs. xtivreg2

    Hi all,

    I discovered that -xtivreg- and -xtivreg2- produces different results with clustered SE, at least in the first stage. Here is what I do:

    xtset pcode year

    xtivreg, fe vce(robust) first

    Code:
    xtivreg growth_GDP_PPP_pc L1.GDP_PPP_pc (HIV_mean_prev= MC_mean_prev) year1-year18, fe vce(robust) first
note: year1 omitted because of collinearity.
note: year18 omitted because of collinearity.
 
First-stage within regression
 
Fixed-effects (within) regression               Number of obs     =      9,360
Group variable: pcode                           Number of groups  =        585
 
R-squared:                                      Obs per group:
     Within  = 0.2537                                         min =         16
     Between = 0.3202                                         avg =       16.0
     Overall = 0.1966                                         max =         16
 
                                                F(17,584)         =      24.34
corr(u_i, Xb) = 0.3450                          Prob > F          =     0.0000
 
                                (Std. err. adjusted for 585 clusters in pcode)
------------------------------------------------------------------------------
             |               Robust
HIV_mean_p~v | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
  GDP_PPP_pc |
         L1. |  -2.00e-09   1.07e-09    -1.87   0.062    -4.11e-09    9.87e-11
             |
       year1 |          0  (omitted)
       year2 |          0  (omitted)
       year3 |   .0156359   .0010809    14.47   0.000      .013513    .0177588
       year4 |   .0147043   .0009866    14.90   0.000     .0127666    .0166421
       year5 |   .0134187   .0009156    14.66   0.000     .0116205    .0152169
       year6 |   .0121423   .0008431    14.40   0.000     .0104864    .0137983
       year7 |   .0108118   .0007856    13.76   0.000     .0092687    .0123548
       year8 |   .0092799   .0007266    12.77   0.000     .0078529    .0107069
       year9 |   .0076829    .000642    11.97   0.000     .0064219    .0089438
      year10 |   .0060961   .0005786    10.54   0.000     .0049597    .0072324
      year11 |   .0050247   .0005304     9.47   0.000      .003983    .0060663
      year12 |   .0042346   .0004855     8.72   0.000     .0032811    .0051881
      year13 |   .0033211   .0004152     8.00   0.000     .0025057    .0041365
      year14 |   .0026812   .0003021     8.87   0.000     .0020878    .0032746
      year15 |   .0020689   .0002121     9.75   0.000     .0016523    .0024856
      year16 |   .0013933   .0001395     9.99   0.000     .0011193    .0016672
      year17 |   .0006147   .0000793     7.76   0.000      .000459    .0007704
      year18 |          0  (omitted)
MC_mean_prev |  -.0144539    .008176    -1.77   0.078    -.0305119    .0016042
       _cons |   .0605409   .0061373     9.86   0.000     .0484871    .0725948
-------------+----------------------------------------------------------------
     sigma_u |  .05936565
     sigma_e |   .0096826
         rho |  .97408738   (fraction of variance due to u_i)
------------------------------------------------------------------------------
 
Fixed-effects (within) IV regression            Number of obs     =      9,945
Group variable: pcode                           Number of groups  =        585
 
R-squared:                                      Obs per group:
     Within  =      .                                         min =         17
     Between = 0.0001                                         avg =       17.0
     Overall = 0.0001                                         max =         17
 
 
                                                Wald chi2(18)     =    5296.46
corr(u_i, Xb) = -0.9752                         Prob > chi2       =     0.0000
 
                                 (Std. err. adjusted for 585 clusters in pcode)
-------------------------------------------------------------------------------
              |               Robust
growth_GDP_~c | Coefficient  std. err.      z    P>|z|     [95% conf. interval]
--------------+----------------------------------------------------------------
HIV_mean_prev |   3.078041    2.36532     1.30   0.193      -1.5579    7.713982
              |
   GDP_PPP_pc |
          L1. |  -1.53e-10   1.10e-08    -0.01   0.989    -2.16e-08    2.13e-08
              |
        year1 |          0  (omitted)
        year2 |  -.0445854   .0410674    -1.09   0.278    -.1250761    .0359053
        year3 |  -.0452986   .0397318    -1.14   0.254    -.1231716    .0325743
        year4 |  -.0404836   .0376005    -1.08   0.282    -.1141792     .033212
        year5 |   .0070452    .034956     0.20   0.840    -.0614673    .0755577
        year6 |  -.0123261   .0313743    -0.39   0.694    -.0738186    .0491664
        year7 |  -.0105576   .0279587    -0.38   0.706    -.0653557    .0442405
        year8 |  -.0025755   .0244086    -0.11   0.916    -.0504155    .0452645
        year9 |  -.0131919   .0207124    -0.64   0.524    -.0537875    .0274037
       year10 |  -.0174418   .0165816    -1.05   0.293    -.0499411    .0150576
       year11 |   .0102819   .0141217     0.73   0.467    -.0173961    .0379599
       year12 |   .0012855   .0120502     0.11   0.915    -.0223324    .0249033
       year13 |   .0033463   .0097785     0.34   0.732    -.0158191    .0225118
       year14 |  -.0055607   .0089031    -0.62   0.532    -.0230104    .0118891
       year15 |   .0062522   .0058038     1.08   0.281     -.005123    .0176273
       year16 |  -.0022079   .0035557    -0.62   0.535     -.009177    .0047611
       year17 |  -.0008809   .0016808    -0.52   0.600    -.0041751    .0024134
       year18 |          0  (omitted)
        _cons |  -.1473728   .1170858    -1.26   0.208    -.3768568    .0821111
--------------+----------------------------------------------------------------
      sigma_u |  .19291119
      sigma_e |  .06111003
          rho |  .90880311   (fraction of variance due to u_i)
-------------------------------------------------------------------------------
Instrumented: HIV_mean_prev
 Instruments: L.GDP_PPP_pc year2 year3 year4 year5 year6 year7 year8 year9
              year10 year11 year12 year13 year14 year15 year16 year17
              MC_mean_prev
r; t=0.89 15:05:03
    xtivreg2, fe cluster(pcode) first

    Code:
     
xtivreg2 growth_GDP_PPP_pc L1.GDP_PPP_pc (HIV_mean_prev= MC_mean_prev) year1-year18, fe cluster(pcode) first
Warning - collinearities detected
Vars dropped:       year1 year18
 
FIXED EFFECTS ESTIMATION
------------------------
Number of groups =       585                    Obs per group: min =        17
                                                               avg =      17.0
                                                               max =        17
Warning - collinearities detected
Vars dropped:  year1 year18
 
First-stage regressions
-----------------------
 
 
FIXED EFFECTS ESTIMATION
------------------------
Number of groups =       585                    Obs per group: min =        17
                                                               avg =      17.0
                                                               max =        17
 
First-stage regression of HIV_mean_prev:
 
Statistics robust to heteroskedasticity and clustering on pcode
Number of obs =                   9945
Number of clusters (pcode) =       585
------------------------------------------------------------------------------
             |               Robust
HIV_mean_p~v | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
MC_mean_prev |  -.0129952   .0089758    -1.45   0.148    -.0305899    .0045994
             |
  GDP_PPP_pc |
         L1. |  -2.99e-09   1.15e-09    -2.61   0.009    -5.23e-09   -7.43e-10
             |
       year1 |          0  (omitted)
       year2 |     .01637   .0011976    13.67   0.000     .0140224    .0187175
       year3 |   .0157388   .0011052    14.24   0.000     .0135723    .0179052
       year4 |   .0148042   .0010155    14.58   0.000     .0128135    .0167948
       year5 |   .0135153   .0009475    14.26   0.000      .011658    .0153726
       year6 |   .0122371   .0008779    13.94   0.000     .0105163    .0139579
       year7 |   .0109056    .000823    13.25   0.000     .0092923    .0125188
       year8 |   .0093755    .000767    12.22   0.000      .007872     .010879
       year9 |   .0077759   .0006847    11.36   0.000     .0064338     .009118
      year10 |    .006185   .0006211     9.96   0.000     .0049675    .0074026
      year11 |   .0051064   .0005723     8.92   0.000     .0039846    .0062282
      year12 |   .0043111   .0005246     8.22   0.000     .0032829    .0053394
      year13 |   .0033855   .0004471     7.57   0.000     .0025091    .0042618
      year14 |   .0027209   .0003192     8.52   0.000     .0020952    .0033467
      year15 |   .0020889   .0002173     9.61   0.000      .001663    .0025148
      year16 |   .0014029   .0001406     9.97   0.000     .0011272    .0016786
      year17 |   .0006228   .0000811     7.68   0.000     .0004638    .0007819
      year18 |          0  (omitted)
------------------------------------------------------------------------------
F test of excluded instruments:
  F(  1,   584) =     2.10
  Prob > F      =   0.1482
Sanderson-Windmeijer multivariate F test of excluded instruments:
  F(  1,   584) =     2.10
  Prob > F      =   0.1482
 
 
 
Summary results for first-stage regressions
-------------------------------------------
 
                                           (Underid)            (Weak id)
Variable     | F(  1,   584)  P-val | SW Chi-sq(  1) P-val | SW F(  1,   584)
HIV_mean_pre |       2.10    0.1482 |        2.10   0.1470 |        2.10
 
NB: first-stage test statistics cluster-robust
 
Stock-Yogo weak ID F test critical values for single endogenous regressor:
                                   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 i.i.d. errors only.
 
Underidentification test
Ho: matrix of reduced form coefficients has rank=K1-1 (underidentified)
Ha: matrix has rank=K1 (identified)
Kleibergen-Paap rk LM statistic          Chi-sq(1)=2.04     P-val=0.1528
 
Weak identification test
Ho: equation is weakly identified
Cragg-Donald Wald F statistic                                      22.54
Kleibergen-Paap Wald rk F statistic                                 2.10
 
Stock-Yogo weak ID test critical values for K1=1 and L1=1:
                                   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.
 
Weak-instrument-robust inference
Tests of joint significance of endogenous regressors B1 in main equation
Ho: B1=0 and orthogonality conditions are valid
Anderson-Rubin Wald test           F(1,584)=      13.11     P-val=0.0003
Anderson-Rubin Wald test           Chi-sq(1)=     13.16     P-val=0.0003
Stock-Wright LM S statistic        Chi-sq(1)=     24.96     P-val=0.0000
 
NB: Underidentification, weak identification and weak-identification-robust
    test statistics cluster-robust
 
Number of clusters             N_clust  =        585
Number of observations               N  =       9945
Number of regressors                 K  =         18
Number of endogenous regressors      K1 =          1
Number of instruments                L  =         18
Number of excluded instruments       L1 =          1
 
IV (2SLS) estimation
--------------------
 
Estimates efficient for homoskedasticity only
Statistics robust to heteroskedasticity and clustering on pcode
 
Number of clusters (pcode) =       585                Number of obs =     9945
                                                      F( 18,   584) =    38.11
                                                      Prob > F      =   0.0000
Total (centered) SS     =  24.53840383                Centered R2   =  -0.4217
Total (uncentered) SS   =  24.53840383                Uncentered R2 =  -0.4217
Residual SS             =   34.8871044                Root MSE      =   .06105
 
-------------------------------------------------------------------------------
              |               Robust
growth_GDP_~c | Coefficient  std. err.      z    P>|z|     [95% conf. interval]
--------------+----------------------------------------------------------------
HIV_mean_prev |   3.078041   2.361157     1.30   0.192    -1.549742    7.705824
              |
   GDP_PPP_pc |
          L1. |  -1.53e-10   1.09e-08    -0.01   0.989    -2.16e-08    2.13e-08
              |
        year1 |          0  (omitted)
        year2 |  -.0445854   .0409952    -1.09   0.277    -.1249344    .0357637
        year3 |  -.0452986   .0396619    -1.14   0.253    -.1230345    .0324373
        year4 |  -.0404836   .0375343    -1.08   0.281    -.1140495    .0330823
        year5 |   .0070452   .0348945     0.20   0.840    -.0613467    .0754371
        year6 |  -.0123261   .0313191    -0.39   0.694    -.0737104    .0490582
        year7 |  -.0105576   .0279095    -0.38   0.705    -.0652593    .0441441
        year8 |  -.0025755   .0243657    -0.11   0.916    -.0503313    .0451803
        year9 |  -.0131919    .020676    -0.64   0.523    -.0537161    .0273323
       year10 |  -.0174418   .0165524    -1.05   0.292    -.0498839    .0150004
       year11 |   .0102819   .0140968     0.73   0.466    -.0173474    .0379111
       year12 |   .0012855   .0120289     0.11   0.915    -.0222908    .0248618
       year13 |   .0033463   .0097613     0.34   0.732    -.0157854     .022478
       year14 |  -.0055607   .0088874    -0.63   0.532    -.0229797    .0118584
       year15 |   .0062522   .0057935     1.08   0.281     -.005103    .0176073
       year16 |  -.0022079   .0035494    -0.62   0.534    -.0091647    .0047488
       year17 |  -.0008809   .0016778    -0.53   0.600    -.0041693    .0024076
       year18 |          0  (omitted)
-------------------------------------------------------------------------------
Underidentification test (Kleibergen-Paap rk LM statistic):              2.044
                                                   Chi-sq(1) P-val =    0.1528
------------------------------------------------------------------------------
Weak identification test (Cragg-Donald Wald F statistic):               22.541
                         (Kleibergen-Paap rk Wald F statistic):          2.096
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.
------------------------------------------------------------------------------
Hansen J statistic (overidentification test of all instruments):         0.000
                                                 (equation exactly identified)
------------------------------------------------------------------------------
Instrumented:         HIV_mean_prev
Included instruments: L.GDP_PPP_pc year2 year3 year4 year5 year6 year7 year8
                      year9 year10 year11 year12 year13 year14 year15 year16
                      year17
Excluded instruments: MC_mean_prev
Dropped collinear:    year1 year18
------------------------------------------------------------------------------
r; t=0.70 15:08:52
    As it was suggested in previous posts on statalist I could add -small- to -xtivreg2-. This gives me:

    xtivreg2, fe cluster(pcode) first small

    Code:
     
xtivreg2 growth_GDP_PPP_pc L1.GDP_PPP_pc (HIV_mean_prev= MC_mean_prev) year1-year18, fe cluster(pcode) first small
Warning - collinearities detected
Vars dropped:       year1 year18
 
FIXED EFFECTS ESTIMATION
------------------------
Number of groups =       585                    Obs per group: min =        17
                                                               avg =      17.0
                                                               max =        17
Warning - collinearities detected
Vars dropped:  year1 year18
 
First-stage regressions
-----------------------
 
 
FIXED EFFECTS ESTIMATION
------------------------
Number of groups =       585                    Obs per group: min =        17
                                                               avg =      17.0
                                                               max =        17
 
First-stage regression of HIV_mean_prev:
 
Statistics robust to heteroskedasticity and clustering on pcode
Number of obs =                   9945
Number of clusters (pcode) =       585
------------------------------------------------------------------------------
             |               Robust
HIV_mean_p~v | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
MC_mean_prev |  -.0129952   .0089758    -1.45   0.148    -.0305899    .0045994
             |
  GDP_PPP_pc |
         L1. |  -2.99e-09   1.15e-09    -2.61   0.009    -5.23e-09   -7.43e-10
             |
       year1 |          0  (omitted)
       year2 |     .01637   .0011976    13.67   0.000     .0140224    .0187175
       year3 |   .0157388   .0011052    14.24   0.000     .0135723    .0179052
       year4 |   .0148042   .0010155    14.58   0.000     .0128135    .0167948
       year5 |   .0135153   .0009475    14.26   0.000      .011658    .0153726
       year6 |   .0122371   .0008779    13.94   0.000     .0105163    .0139579
       year7 |   .0109056    .000823    13.25   0.000     .0092923    .0125188
       year8 |   .0093755    .000767    12.22   0.000      .007872     .010879
       year9 |   .0077759   .0006847    11.36   0.000     .0064338     .009118
      year10 |    .006185   .0006211     9.96   0.000     .0049675    .0074026
      year11 |   .0051064   .0005723     8.92   0.000     .0039846    .0062282
      year12 |   .0043111   .0005246     8.22   0.000     .0032829    .0053394
      year13 |   .0033855   .0004471     7.57   0.000     .0025091    .0042618
      year14 |   .0027209   .0003192     8.52   0.000     .0020952    .0033467
      year15 |   .0020889   .0002173     9.61   0.000      .001663    .0025148
      year16 |   .0014029   .0001406     9.97   0.000     .0011272    .0016786
      year17 |   .0006228   .0000811     7.68   0.000     .0004638    .0007819
      year18 |          0  (omitted)
------------------------------------------------------------------------------
F test of excluded instruments:
  F(  1,   584) =     2.10
  Prob > F      =   0.1482
Sanderson-Windmeijer multivariate F test of excluded instruments:
  F(  1,   584) =     2.10
  Prob > F      =   0.1482
 
 
 
Summary results for first-stage regressions
-------------------------------------------
 
                                           (Underid)            (Weak id)
Variable     | F(  1,   584)  P-val | SW Chi-sq(  1) P-val | SW F(  1,   584)
HIV_mean_pre |       2.10    0.1482 |        2.10   0.1470 |        2.10
 
NB: first-stage test statistics cluster-robust
 
Stock-Yogo weak ID F test critical values for single endogenous regressor:
                                   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 i.i.d. errors only.
 
Underidentification test
Ho: matrix of reduced form coefficients has rank=K1-1 (underidentified)
Ha: matrix has rank=K1 (identified)
Kleibergen-Paap rk LM statistic          Chi-sq(1)=2.04     P-val=0.1528
 
Weak identification test
Ho: equation is weakly identified
Cragg-Donald Wald F statistic                                      22.54
Kleibergen-Paap Wald rk F statistic                                 2.10
 
Stock-Yogo weak ID test critical values for K1=1 and L1=1:
                                   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.
 
Weak-instrument-robust inference
Tests of joint significance of endogenous regressors B1 in main equation
Ho: B1=0 and orthogonality conditions are valid
Anderson-Rubin Wald test           F(1,584)=      13.11     P-val=0.0003
Anderson-Rubin Wald test           Chi-sq(1)=     13.16     P-val=0.0003
Stock-Wright LM S statistic        Chi-sq(1)=     24.96     P-val=0.0000
 
NB: Underidentification, weak identification and weak-identification-robust
    test statistics cluster-robust
 
Number of clusters             N_clust  =        585
Number of observations               N  =       9945
Number of regressors                 K  =         18
Number of endogenous regressors      K1 =          1
Number of instruments                L  =         18
Number of excluded instruments       L1 =          1
 
IV (2SLS) estimation
--------------------
 
Estimates efficient for homoskedasticity only
Statistics robust to heteroskedasticity and clustering on pcode
 
Number of clusters (pcode) =       585                Number of obs =     9945
                                                      F( 18,   584) =    38.11
                                                      Prob > F      =   0.0000
Total (centered) SS     =  24.53840383                Centered R2   =  -0.4217
Total (uncentered) SS   =  24.53840383                Uncentered R2 =  -0.4217
Residual SS             =   34.8871044                Root MSE      =   .06111
 
-------------------------------------------------------------------------------
              |               Robust
growth_GDP_~c | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
--------------+----------------------------------------------------------------
HIV_mean_prev |   3.078041     2.3652     1.30   0.194    -1.567294    7.723376
              |
   GDP_PPP_pc |
          L1. |  -1.53e-10   1.10e-08    -0.01   0.989    -2.17e-08    2.14e-08
              |
        year1 |          0  (omitted)
        year2 |  -.0445854   .0410654    -1.09   0.278    -.1252391    .0360684
        year3 |  -.0452986   .0397298    -1.14   0.255    -.1233294    .0327321
        year4 |  -.0404836   .0375986    -1.08   0.282    -.1143285    .0333613
        year5 |   .0070452   .0349542     0.20   0.840    -.0616061    .0756965
        year6 |  -.0123261   .0313727    -0.39   0.695    -.0739432     .049291
        year7 |  -.0105576   .0279573    -0.38   0.706    -.0654667    .0443515
        year8 |  -.0025755   .0244074    -0.11   0.916    -.0505124    .0453614
        year9 |  -.0131919   .0207114    -0.64   0.524    -.0538698     .027486
       year10 |  -.0174418   .0165808    -1.05   0.293    -.0500069    .0151234
       year11 |   .0102819    .014121     0.73   0.467    -.0174522    .0380159
       year12 |   .0012855   .0120495     0.11   0.915    -.0223803    .0249512
       year13 |   .0033463    .009778     0.34   0.732     -.015858    .0225506
       year14 |  -.0055607   .0089026    -0.62   0.532    -.0230457    .0119244
       year15 |   .0062522   .0058035     1.08   0.282     -.005146    .0176504
       year16 |  -.0022079   .0035555    -0.62   0.535    -.0091911    .0047752
       year17 |  -.0008809   .0016807    -0.52   0.600    -.0041818      .00242
       year18 |          0  (omitted)
-------------------------------------------------------------------------------
Underidentification test (Kleibergen-Paap rk LM statistic):              2.044
                                                   Chi-sq(1) P-val =    0.1528
------------------------------------------------------------------------------
Weak identification test (Cragg-Donald Wald F statistic):               22.541
                         (Kleibergen-Paap rk Wald F statistic):          2.096
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.
------------------------------------------------------------------------------
Hansen J statistic (overidentification test of all instruments):         0.000
                                                 (equation exactly identified)
------------------------------------------------------------------------------
Instrumented:         HIV_mean_prev
Included instruments: L.GDP_PPP_pc year2 year3 year4 year5 year6 year7 year8
                      year9 year10 year11 year12 year13 year14 year15 year16
                      year17
Excluded instruments: MC_mean_prev
Dropped collinear:    year1 year18
------------------------------------------------------------------------------
r; t=0.66 15:11:40
    Can someone explain me why at least the first stage is significantly different and thus the relevance of my instrument is totally different?
    Tags: None
  • #2
    xtivreg2 is from SSC, as you are asked to explain (FAQ Advice #12). You should provide a reproducible example. Those first stage regressions are not equivalent. If you look at the number of observations in xtivreg's first stage, it is 9360 compared to 9945 for xtivreg2's first stage. You will normally want to use the same sample in the first and second stage as xtivreg2 does, so it may be a bug in xtivreg. You need a reproducible example to document it.
    • 1 like

    Comment

    • #3
      Andrew Musau Thank you! Here is a reproducible example:

      Code:
      * Example generated by -dataex-. For more info, type help dataex
clear
input float(growth_GDP_PPP_pc GDP_PPP_pc) double(HIV_mean_prev MC_mean_prev) byte(year1 year2 year3 year4 year5 year6 year7 year8 year9 year10 year11 year12 year13 year14 year15 year16 year17 year18) float(year pcode)
           .  6833.586  .009391321464879853  .874858446015513 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2000 11
  .065409355  7280.567  .010551745858990144 .8882725118842353 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2001 11
   -.0510628  6908.801  .011614275863107404 .8989287794950727 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2002 11
   .07137197  7401.896  .012688413908386263 .9093764177339891 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2003 11
   .12782831  8348.067   .01332622815010693 .9207094649630526 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 2004 11
   .20275557 10040.685  .013409183981135634 .9300142549876839 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 2005 11
     .204221 12091.203   .01390926253078818 .9391515190886699 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 2006 11
   .22216024 14777.388  .014479589038941403 .9468483830665041 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 2007 11
    .1340643 16758.508  .015268043919438909 .9537642038793109 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 2008 11
  .019763516 17089.715  .015508624518019482 .9590509581896545 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 2009 11
  .028940905 17584.307  .015607664117869449 .9633054620073891 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 2010 11
-.0007541792 17571.045  .015496642564496788 .9678465560344831 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 2011 11
   .01161579 17775.146  .015328731367208541 .9708436201970425 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 2012 11
  .024825085 18216.416  .015214403489090759 .9739637299876844 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 2013 11
  .009984556 18398.299   .01591842509966729  .976787445258622 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 2014 11
 -.006766081 18273.814  .016190094392164036 .9793984546798032 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 2015 11
 -.006423553 18156.432  .016184303744952996 .9842327139162554 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 2016 11
 -.006106651 18045.557  .015725322049386394 .9853705811162887 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 2017 11
           .  5603.201  .005378663071760066 .8244465899051402 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2000 12
   .05011557  5884.009  .006095451145050286 .8360699947646738 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2001 12
   -.0650277  5501.385  .006981089430165518 .8501499220518246 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2002 12
   .05520209  5805.073  .007988266687043496 .8619590320465341 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2003 12
    .1103648  6445.749  .009091002163573381 .8745047025383393 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 2004 12
   .18364127  7629.455  .010366657063021837 .8858756469063971 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 2005 12
    .2078017  9214.868  .011385867183695074 .8966687141723947 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 2006 12
   .22580376  11295.62  .012622666238551261 .9059945566895828 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 2007 12
   .13745415  12848.25   .01357129606915773 .9134907837123196 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 2008 12
  .022820244  13141.45  .014793703333036397 .9205730870438915 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 2009 12
   .03203381  13562.42  .015241338952275387 .9276793374405083 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 2010 12
   .03842044 14083.495  .015550171011785718 .9343819989423587 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 2011 12
   .05012392 14789.415  .015946984491737478 .9405230260179795 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 2012 12
  .012813944 14978.926  .016403963430386418 .9470865868323658 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 2013 12
   .04595095 15667.222   .01586490968718549 .9521253695399271 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 2014 12
  .027682135 16100.924  .015224267573324127  .958218850449497 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 2015 12
  .027169807 16538.383  .015239616436908458  .971725254944475 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 2016 12
   .02668107 16979.645  .015207947335857256 .9743931975899267 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 2017 12
           . 10231.248 .0053579459659802795 .8477491841230431 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2000 13
   .08185945 11068.772   .00610771525478817 .8596572050203141 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2001 13
  -.03563569  10674.33  .007212398098500632 .8715866559199094 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2002 13
   .08974355  11632.28  .008368264599441098 .8842979902495656 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2003 13
   .14826533 13356.945   .00972102690816198 .8956824537724906 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 2004 13
    .2258328 16373.382  .011062183202116237 .9076131478816003 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 2005 13
   .24431905  20373.71   .01286353321706326 .9184203784677883 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 2006 13
   .26434293  25759.36  .015083519097047875 .9280267872896119 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 2007 13
   .17468932  30259.24  .016622429230648786 .9346627101276833 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 2008 13
   .05773101  32006.14  .018557102440246144 .9399498089669179 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 2009 13
   .06881657  34208.69  .020675375221164177 .9457677234184559 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 2010 13
   .06981379  36596.93  .022730767563930256 .9510100616076619 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 2011 13
   .08313336  39639.36   .02446623672787976 .9561014218804393 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 2012 13
   .10105586  43645.14    .0261302774492165 .9620655884793992 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 2013 13
    .0814095  47198.27  .026687361430257853 .9666301069936155 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 2014 13
     .064066  50222.08  .027227779871714423 .9709495481137532 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 2015 13
   .06520881     53497   .02544559363586027 .9672400532211274 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 2016 13
   .06652096  57055.67  .025062884862447676 .9712222922491737 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 2017 13
           . 19735.164    .0158507426063955 .9088661339347817 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2000 14
   -.0953218 17853.973  .017563566012084586 .9212442178247737 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2001 14
   -.1770627   14692.7    .0210045430010179 .9327386120845921 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2002 14
  -.06157534 13787.992  .025167927764963887 .9431070927492452 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2003 14
-.0024372996 13754.387   .02732744144438977 .9531785818731122 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 2004 14
   .07188154 14743.073  .027765859514925054  .961212710271903 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 2005 14
  .034295037 15248.688   .02668508274924473 .9686173658610276 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 2006 14
   .06888759 16299.133    .0261502855336378 .9755644166163139 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 2007 14
 .0084982235 16437.646  .026555359881269967 .9846690691842898 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 2008 14
  -.08117697 15103.288   .02556999056428401 .9904222782477345 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 2009 14
  -.06390372 14138.132  .023941940152483045  .991399269184291 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 2010 14
  -.10997305 12583.318   .02231318695510082 .9926917870090635 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 2011 14
   -.0826145 11543.754   .01894537352441931 .9916409081570994 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 2012 14
  -.10236596 10362.066  .015905482173978924 .9906466851963744 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 2013 14
   -.0596228  9744.251  .014709798996182006 .9899813072507547 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 2014 14
  -.06737326  9087.749  .013288223125558244 .9901024740181263 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 2015 14
  -.06044485  8538.441  .013347139182497728 .9926799619335359 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 2016 14
  -.05462177  8072.057  .012906610170868858 .9928491021931352 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 2017 14
           .  2750.393  .018801156639609844  .838788111184522 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2000 15
   .07501148  2956.704   .02316286753288652  .854237033716491 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2001 15
 -.033410236   2857.92  .029025099882279935 .8668547395793825 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2002 15
   .08392139   3097.76  .032881023081441974 .8787375764752527 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2003 15
   .14311618    3541.1   .03714578186987525 .8888767138195899 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 2004 15
    .2084594 4279.2754   .03990454076155636 .8983239109484562 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 2005 15
   .23020905  5264.403   .04131915732474206 .9087265082278356 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 2006 15
    .2491253  6575.899  .044030783905350056 .9182325202216437 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 2007 15
   .16273306  7646.016  .044003059103624106 .9256053751597919 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 2008 15
    .0504075  8031.432   .04732585450980006 .9329395385762861 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 2009 15
   .06284044  8536.131  .049391370331004415 .9401638024329884 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 2010 15
   .06401706  9082.589   .05066468884880395 .9456142115566992 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 2011 15
   .07623564  9775.006   .05608837323820161 .9503570950927802 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 2012 15
    .0930572  10684.64   .05887464340309269 .9542986098453653 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 2013 15
    .0746351  11482.09    .0587530201356666 .9584097115463882 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 2014 15
   .05735087 12140.598   .05789915029209146 .9621552758865981 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 2015 15
   .05783475 12842.746    .0563306781812811 .9642798777113373 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 2016 15
   .05842024 13593.022   .05524786703518986 .9695802650224302 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 2017 15
           .  7219.191  .007577533332232794 .9082650669682574 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2000 16
    .0952809  7907.042  .008516541805653689 .9212861749999998 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2001 16
 -.022636257  7728.056  .009513770641476021 .9313484075971735 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2002 16
   .10576286  8545.397  .010659253766082306  .940787364045936 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2003 16
   .16674684  9970.315  .011575857055613726 .9498460453180192 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 2004 16
   .24753077 12438.275  .011834293735532794 .9581848642226142 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 2005 16
   .25947002 15665.635  .012520292862190802 .9654517273851576 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 2006 16
   .28106532   20068.7   .01314457409765058 .9717857459363946 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 2007 16
    .1916271  23914.41  .013776092365542066 .9766917868374545 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 2008 16
    .0744237 25694.207  .013797905227801246 .9802507946113072 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 2009 16
end
      with xtset pcode year


      PS: It seems that this problem occurs mainly due to clustered standard errors
      Last edited by Michael Schuster; 03 Feb 2022, 09:42.

      Comment

      • #4
        Thanks for the data example. I think it is an oversight from the Stata developers as it relates to reproducing the first stage results. I would send a report to Technical Services for them to look into it. Otherwise, I would put more weight on xtivreg2's first stage results or reproduce xtivreg's first stage using xtreg as below:


        Code:
        * Example generated by -dataex-. For more info, type help dataex
clear
input float(growth_GDP_PPP_pc GDP_PPP_pc) double(HIV_mean_prev MC_mean_prev) byte(year1 year2 year3 year4 year5 year6 year7 year8 year9 year10 year11 year12 year13 year14 year15 year16 year17 year18) float(year pcode sample)
           .  6833.586  .009391321464879853  .874858446015513 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2000 11 0
  .065409355  7280.567  .010551745858990144 .8882725118842353 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2001 11 1
   -.0510628  6908.801  .011614275863107404 .8989287794950727 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2002 11 1
   .07137197  7401.896  .012688413908386263 .9093764177339891 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2003 11 1
   .12782831  8348.067   .01332622815010693 .9207094649630526 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 2004 11 1
   .20275557 10040.685  .013409183981135634 .9300142549876839 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 2005 11 1
     .204221 12091.203   .01390926253078818 .9391515190886699 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 2006 11 1
   .22216024 14777.388  .014479589038941403 .9468483830665041 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 2007 11 1
    .1340643 16758.508  .015268043919438909 .9537642038793109 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 2008 11 1
  .019763516 17089.715  .015508624518019482 .9590509581896545 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 2009 11 1
  .028940905 17584.307  .015607664117869449 .9633054620073891 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 2010 11 1
-.0007541792 17571.045  .015496642564496788 .9678465560344831 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 2011 11 1
   .01161579 17775.146  .015328731367208541 .9708436201970425 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 2012 11 1
  .024825085 18216.416  .015214403489090759 .9739637299876844 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 2013 11 1
  .009984556 18398.299   .01591842509966729  .976787445258622 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 2014 11 1
 -.006766081 18273.814  .016190094392164036 .9793984546798032 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 2015 11 1
 -.006423553 18156.432  .016184303744952996 .9842327139162554 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 2016 11 1
 -.006106651 18045.557  .015725322049386394 .9853705811162887 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 2017 11 1
           .  5603.201  .005378663071760066 .8244465899051402 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2000 12 0
   .05011557  5884.009  .006095451145050286 .8360699947646738 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2001 12 1
   -.0650277  5501.385  .006981089430165518 .8501499220518246 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2002 12 1
   .05520209  5805.073  .007988266687043496 .8619590320465341 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2003 12 1
    .1103648  6445.749  .009091002163573381 .8745047025383393 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 2004 12 1
   .18364127  7629.455  .010366657063021837 .8858756469063971 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 2005 12 1
    .2078017  9214.868  .011385867183695074 .8966687141723947 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 2006 12 1
   .22580376  11295.62  .012622666238551261 .9059945566895828 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 2007 12 1
   .13745415  12848.25   .01357129606915773 .9134907837123196 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 2008 12 1
  .022820244  13141.45  .014793703333036397 .9205730870438915 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 2009 12 1
   .03203381  13562.42  .015241338952275387 .9276793374405083 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 2010 12 1
   .03842044 14083.495  .015550171011785718 .9343819989423587 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 2011 12 1
   .05012392 14789.415  .015946984491737478 .9405230260179795 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 2012 12 1
  .012813944 14978.926  .016403963430386418 .9470865868323658 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 2013 12 1
   .04595095 15667.222   .01586490968718549 .9521253695399271 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 2014 12 1
  .027682135 16100.924  .015224267573324127  .958218850449497 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 2015 12 1
  .027169807 16538.383  .015239616436908458  .971725254944475 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 2016 12 1
   .02668107 16979.645  .015207947335857256 .9743931975899267 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 2017 12 1
           . 10231.248 .0053579459659802795 .8477491841230431 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2000 13 0
   .08185945 11068.772   .00610771525478817 .8596572050203141 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2001 13 1
  -.03563569  10674.33  .007212398098500632 .8715866559199094 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2002 13 1
   .08974355  11632.28  .008368264599441098 .8842979902495656 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2003 13 1
   .14826533 13356.945   .00972102690816198 .8956824537724906 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 2004 13 1
    .2258328 16373.382  .011062183202116237 .9076131478816003 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 2005 13 1
   .24431905  20373.71   .01286353321706326 .9184203784677883 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 2006 13 1
   .26434293  25759.36  .015083519097047875 .9280267872896119 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 2007 13 1
   .17468932  30259.24  .016622429230648786 .9346627101276833 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 2008 13 1
   .05773101  32006.14  .018557102440246144 .9399498089669179 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 2009 13 1
   .06881657  34208.69  .020675375221164177 .9457677234184559 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 2010 13 1
   .06981379  36596.93  .022730767563930256 .9510100616076619 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 2011 13 1
   .08313336  39639.36   .02446623672787976 .9561014218804393 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 2012 13 1
   .10105586  43645.14    .0261302774492165 .9620655884793992 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 2013 13 1
    .0814095  47198.27  .026687361430257853 .9666301069936155 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 2014 13 1
     .064066  50222.08  .027227779871714423 .9709495481137532 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 2015 13 1
   .06520881     53497   .02544559363586027 .9672400532211274 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 2016 13 1
   .06652096  57055.67  .025062884862447676 .9712222922491737 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 2017 13 1
           . 19735.164    .0158507426063955 .9088661339347817 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2000 14 0
   -.0953218 17853.973  .017563566012084586 .9212442178247737 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2001 14 1
   -.1770627   14692.7    .0210045430010179 .9327386120845921 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2002 14 1
  -.06157534 13787.992  .025167927764963887 .9431070927492452 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2003 14 1
-.0024372996 13754.387   .02732744144438977 .9531785818731122 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 2004 14 1
   .07188154 14743.073  .027765859514925054  .961212710271903 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 2005 14 1
  .034295037 15248.688   .02668508274924473 .9686173658610276 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 2006 14 1
   .06888759 16299.133    .0261502855336378 .9755644166163139 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 2007 14 1
 .0084982235 16437.646  .026555359881269967 .9846690691842898 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 2008 14 1
  -.08117697 15103.288   .02556999056428401 .9904222782477345 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 2009 14 1
  -.06390372 14138.132  .023941940152483045  .991399269184291 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 2010 14 1
  -.10997305 12583.318   .02231318695510082 .9926917870090635 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 2011 14 1
   -.0826145 11543.754   .01894537352441931 .9916409081570994 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 2012 14 1
  -.10236596 10362.066  .015905482173978924 .9906466851963744 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 2013 14 1
   -.0596228  9744.251  .014709798996182006 .9899813072507547 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 2014 14 1
  -.06737326  9087.749  .013288223125558244 .9901024740181263 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 2015 14 1
  -.06044485  8538.441  .013347139182497728 .9926799619335359 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 2016 14 1
  -.05462177  8072.057  .012906610170868858 .9928491021931352 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 2017 14 1
           .  2750.393  .018801156639609844  .838788111184522 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2000 15 0
   .07501148  2956.704   .02316286753288652  .854237033716491 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2001 15 1
 -.033410236   2857.92  .029025099882279935 .8668547395793825 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2002 15 1
   .08392139   3097.76  .032881023081441974 .8787375764752527 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2003 15 1
   .14311618    3541.1   .03714578186987525 .8888767138195899 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 2004 15 1
    .2084594 4279.2754   .03990454076155636 .8983239109484562 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 2005 15 1
   .23020905  5264.403   .04131915732474206 .9087265082278356 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 2006 15 1
    .2491253  6575.899  .044030783905350056 .9182325202216437 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 2007 15 1
   .16273306  7646.016  .044003059103624106 .9256053751597919 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 2008 15 1
    .0504075  8031.432   .04732585450980006 .9329395385762861 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 2009 15 1
   .06284044  8536.131  .049391370331004415 .9401638024329884 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 2010 15 1
   .06401706  9082.589   .05066468884880395 .9456142115566992 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 2011 15 1
   .07623564  9775.006   .05608837323820161 .9503570950927802 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 2012 15 1
    .0930572  10684.64   .05887464340309269 .9542986098453653 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 2013 15 1
    .0746351  11482.09    .0587530201356666 .9584097115463882 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 2014 15 1
   .05735087 12140.598   .05789915029209146 .9621552758865981 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 2015 15 1
   .05783475 12842.746    .0563306781812811 .9642798777113373 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 2016 15 1
   .05842024 13593.022   .05524786703518986 .9695802650224302 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 2017 15 1
           .  7219.191  .007577533332232794 .9082650669682574 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2000 16 0
    .0952809  7907.042  .008516541805653689 .9212861749999998 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2001 16 1
 -.022636257  7728.056  .009513770641476021 .9313484075971735 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2002 16 1
   .10576286  8545.397  .010659253766082306  .940787364045936 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2003 16 1
   .16674684  9970.315  .011575857055613726 .9498460453180192 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 2004 16 1
   .24753077 12438.275  .011834293735532794 .9581848642226142 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 2005 16 1
   .25947002 15665.635  .012520292862190802 .9654517273851576 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 2006 16 1
   .28106532   20068.7   .01314457409765058 .9717857459363946 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 2007 16 1
    .1916271  23914.41  .013776092365542066 .9766917868374545 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 2008 16 1
    .0744237 25694.207  .013797905227801246 .9802507946113072 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 2009 16 1
end

xtset pcode year
xtivreg2 growth_GDP_PPP_pc L1.GDP_PPP_pc (HIV_mean_prev= MC_mean_prev) year1-year18, fe cluster(pcode) first
xtivreg growth_GDP_PPP_pc L1.GDP_PPP_pc (HIV_mean_prev= MC_mean_prev) year1-year18, fe vce(robust) first
xtreg HIV_mean_prev MC_mean_prev L.GDP_PPP_pc i.year if e(sample), fe robust
        Res.:

        Code:
        . xtivreg2 growth_GDP_PPP_pc L1.GDP_PPP_pc (HIV_mean_prev= MC_mean_prev) year1-year18, fe cluster(pcode) first
Warning - collinearities detected
Vars dropped:       year1 year18

FIXED EFFECTS ESTIMATION
------------------------
Number of groups =         6                    Obs per group: min =         9
                                                               avg =      15.7
                                                               max =        17
Warning - collinearities detected
Vars dropped:  year1 year18

First-stage regressions
-----------------------


FIXED EFFECTS ESTIMATION
------------------------
Number of groups =         6                    Obs per group: min =         9
                                                               avg =      15.7
                                                               max =        17

First-stage regression of HIV_mean_prev:

Statistics robust to heteroskedasticity and clustering on pcode
Number of obs =                     94
Number of clusters (pcode) =         6
------------------------------------------------------------------------------
             |               Robust
HIV_mean_p~v |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
MC_mean_prev |   .3595118   .2454461     1.46   0.147    -.1300151    .8490387
             |
  GDP_PPP_pc |
         L1. |   2.96e-07   1.96e-07     1.51   0.136    -9.53e-08    6.88e-07
             |
       year1 |          0  (omitted)
       year2 |    .030363   .0188799     1.61   0.112    -.0072917    .0680178
       year3 |   .0283154   .0167781     1.69   0.096    -.0051474    .0617782
       year4 |    .026615    .014507     1.83   0.071    -.0023182    .0555482
       year5 |   .0243931   .0124666     1.96   0.054    -.0004707    .0492569
       year6 |   .0216639   .0104902     2.07   0.043     .0007418     .042586
       year7 |   .0185447   .0085187     2.18   0.033     .0015548    .0355347
       year8 |   .0161114   .0069469     2.32   0.023     .0022563    .0299665
       year9 |   .0134473   .0055728     2.41   0.018     .0023326     .024562
      year10 |   .0117047   .0046143     2.54   0.013     .0025018    .0209077
      year11 |   .0105712   .0032613     3.24   0.002     .0040668    .0170757
      year12 |    .009123   .0024539     3.72   0.000     .0042289    .0140171
      year13 |   .0085265   .0024539     3.47   0.001     .0036324    .0134207
      year14 |   .0073264   .0023793     3.08   0.003      .002581    .0120718
      year15 |   .0058073   .0018499     3.14   0.002     .0021177    .0094968
      year16 |   .0038991   .0013054     2.99   0.004     .0012956    .0065026
      year17 |   .0016548   .0009002     1.84   0.070    -.0001405    .0034501
      year18 |          0  (omitted)
------------------------------------------------------------------------------
F test of excluded instruments:
  F(  1,     5) =     2.15
  Prob > F      =   0.2029
Sanderson-Windmeijer multivariate F test of excluded instruments:
  F(  1,     5) =     2.15
  Prob > F      =   0.2029



Summary results for first-stage regressions
-------------------------------------------

                                           (Underid)            (Weak id)
Variable     | F(  1,     5)  P-val | SW Chi-sq(  1) P-val | SW F(  1,     5)
HIV_mean_pre |       2.15    0.2029 |        3.15   0.0759 |        2.15

NB: first-stage test statistics cluster-robust

Stock-Yogo weak ID F test critical values for single endogenous regressor:
                                   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 i.i.d. errors only.

Underidentification test
Ho: matrix of reduced form coefficients has rank=K1-1 (underidentified)
Ha: matrix has rank=K1 (identified)
Kleibergen-Paap rk LM statistic          Chi-sq(1)=3.16     P-val=0.0752

Weak identification test
Ho: equation is weakly identified
Cragg-Donald Wald F statistic                                      32.24
Kleibergen-Paap Wald rk F statistic                                 2.15

Stock-Yogo weak ID test critical values for K1=1 and L1=1:
                                   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.

Weak-instrument-robust inference
Tests of joint significance of endogenous regressors B1 in main equation
Ho: B1=0 and orthogonality conditions are valid
Anderson-Rubin Wald test           F(1,5)=         0.27     P-val=0.6281
Anderson-Rubin Wald test           Chi-sq(1)=      0.39     P-val=0.5321
Stock-Wright LM S statistic        Chi-sq(1)=         .     P-val=     .

NB: Underidentification, weak identification and weak-identification-robust
    test statistics cluster-robust

Number of clusters             N_clust  =          6
Number of observations               N  =         94
Number of regressors                 K  =         18
Number of endogenous regressors      K1 =          1
Number of instruments                L  =         18
Number of excluded instruments       L1 =          1

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

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

Number of clusters (pcode) =         6                Number of obs =       94
                                                      F( 18,     5) =   434.26
                                                      Prob > F      =   0.0000
Total (centered) SS     =  .5454066095                Centered R2   =   0.9667
Total (uncentered) SS   =  .5454066095                Uncentered R2 =   0.9667
Residual SS             =  .0181448336                Root MSE      =   .01436

-------------------------------------------------------------------------------
              |               Robust
growth_GDP_~c |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
--------------+----------------------------------------------------------------
HIV_mean_prev |  -.5903933   .9577088    -0.62   0.538    -2.467468    1.286681
              |
   GDP_PPP_pc |
          L1. |  -1.08e-07   5.31e-07    -0.20   0.839    -1.15e-06    9.33e-07
              |
        year1 |          0  (omitted)
        year2 |   .0082763   .0223056     0.37   0.711    -.0354419    .0519944
        year3 |  -.0999311   .0162518    -6.15   0.000    -.1317841   -.0680781
        year4 |   .0227503   .0152115     1.50   0.135    -.0070636    .0525642
        year5 |   .0820543   .0134599     6.10   0.000     .0556734    .1084351
        year6 |   .1571221   .0133681    11.75   0.000     .1309212    .1833231
        year7 |   .1644332   .0180202     9.12   0.000     .1291143    .1997521
        year8 |   .1871723   .0151091    12.39   0.000     .1575591    .2167855
        year9 |   .1041786   .0109532     9.51   0.000     .0827107    .1256464
       year10 |   -.005869   .0077553    -0.76   0.449    -.0210692    .0093312
       year11 |   .0071277   .0053965     1.32   0.187    -.0034493    .0177046
       year12 |  -.0060316   .0097924    -0.62   0.538    -.0252245    .0131612
       year13 |   .0098779   .0073248     1.35   0.177    -.0044785    .0242342
       year14 |   .0083411   .0133941     0.62   0.533    -.0179108     .034593
       year15 |   .0129593   .0033855     3.83   0.000     .0063238    .0195948
       year16 |  -.0026692   .0018893    -1.41   0.158    -.0063721    .0010336
       year17 |  -.0013077   .0008296    -1.58   0.115    -.0029337    .0003183
       year18 |          0  (omitted)
-------------------------------------------------------------------------------
Underidentification test (Kleibergen-Paap rk LM statistic):              3.165
                                                   Chi-sq(1) P-val =    0.0752
------------------------------------------------------------------------------
Weak identification test (Cragg-Donald Wald F statistic):               32.239
                         (Kleibergen-Paap rk Wald F statistic):          2.145
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:         HIV_mean_prev
Included instruments: L.GDP_PPP_pc year2 year3 year4 year5 year6 year7 year8
                      year9 year10 year11 year12 year13 year14 year15 year16
                      year17
Excluded instruments: MC_mean_prev
Dropped collinear:    year1 year18
------------------------------------------------------------------------------



.
. xtreg HIV_mean_prev MC_mean_prev L.GDP_PPP_pc i.year if e(sample), fe robust

Fixed-effects (within) regression               Number of obs     =         94
Group variable: pcode                           Number of groups  =          6

R-sq:                                           Obs per group:
     within  = 0.7090                                         min =          9
     between = 0.1592                                         avg =       15.7
     overall = 0.0039                                         max =         17

                                                F(5,5)            =          .
corr(u_i, Xb)  = -0.6370                        Prob > F          =          .

                                  (Std. Err. adjusted for 6 clusters in pcode)
------------------------------------------------------------------------------
             |               Robust
HIV_mean_p~v |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
MC_mean_prev |   .3595118    .247077     1.46   0.205    -.2756199    .9946435
             |
  GDP_PPP_pc |
         L1. |   2.96e-07   1.98e-07     1.50   0.194    -2.12e-07    8.05e-07
             |
        year |
       2002  |  -.0020477   .0021716    -0.94   0.389    -.0076299    .0035345
       2003  |  -.0037481   .0046281    -0.81   0.455     -.015645    .0081489
       2004  |    -.00597   .0067246    -0.89   0.415     -.023256    .0113161
       2005  |  -.0086992   .0087366    -1.00   0.365    -.0311572    .0137589
       2006  |  -.0118183   .0106498    -1.11   0.318    -.0391946     .015558
       2007  |  -.0142516   .0121855    -1.17   0.295    -.0455753    .0170721
       2008  |  -.0169158   .0135221    -1.25   0.266    -.0516754    .0178439
       2009  |  -.0186583   .0144458    -1.29   0.253    -.0557924    .0184758
       2010  |  -.0197918   .0158215    -1.25   0.266    -.0604623    .0208787
       2011  |    -.02124   .0166797    -1.27   0.259    -.0641165    .0216365
       2012  |  -.0218365   .0169666    -1.29   0.254    -.0654506    .0217776
       2013  |  -.0230367   .0174418    -1.32   0.244    -.0678721    .0217988
       2014  |  -.0245558   .0177958    -1.38   0.226    -.0703014    .0211898
       2015  |   -.026464    .018185    -1.46   0.205      -.07321     .020282
       2016  |  -.0287082    .018337    -1.57   0.178    -.0758449    .0184284
       2017  |   -.030363   .0190053    -1.60   0.171    -.0792178    .0184917
             |
       _cons |  -.3053648   .2178495    -1.40   0.220    -.8653649    .2546352
-------------+----------------------------------------------------------------
     sigma_u |  .01877582
     sigma_e |  .00386111
         rho |  .95942689   (fraction of variance due to u_i)
------------------------------------------------------------------------------

        Comment

        Previous Next
        Working...
        X