Find first stage F-stats under xtivreg with factor variables (so no xtoverid)

[Leah Bevis]

It seems to be well-documented (here: https://www.statalist.org/forums/for...port-fvvarlist or here: https://www.stata.com/statalist/arch.../msg00707.html) that xtoverid does not work when factor variable are included in a regression using xtivreg.

I am using factor variables in an xtivreg regression, and I would like to know the first stage F-stat for my excluded variables. Is there any way to do this w/out using xtoverid?

If there is no post-estimation command that works to do this, I can of course separately run what I think is the 1st stage, and test my excluded variables myself. From page 20 of the manual (https://www.stata.com/manuals/xtxtivreg.pdf) it looks like I would first (a) remove all fixed effects using xtreg, then (b) run a 2SLS regression of my 1st stage using ivreg or ivreg2. Does anyone know if this is indeed the best manual approximation of the first stage of xtivreg?

[Andrew Musau]

No need to program this for fixed effects IV regression. Install ivreghdfe from SSC which supports factor variables. The test is part of the default output.

webuse nlswork
xtivreg ln_w age not_smsa c.age#c.age (tenure = union south), fe
ivreghdfe ln_w age not_smsa c.age#c.age (tenure = union south), a(idcode)

Res.:

. ivreghdfe ln_w age not_smsa c.age#c.age (tenure = union south), a(idcode)
(dropped 673 singleton observations)
(MWFE estimator converged in 1 iterations)

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

Estimates efficient for homoskedasticity only
Statistics consistent for homoskedasticity only

                                                      Number of obs =    18334
                                                      F(  4, 14869) =    74.14
                                                      Prob > F      =   0.0000
Total (centered) SS     =  1119.473248                Centered R2   =  -4.2766
Total (uncentered) SS   =  1119.473248                Uncentered R2 =  -4.2766
Residual SS             =  5907.007821                Root MSE      =    .6303

------------------------------------------------------------------------------
     ln_wage |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
      tenure |   .2403531   .0373419     6.44   0.000     .1671583    .3135478
         age |   .0118437   .0090032     1.32   0.188    -.0058037    .0294912
    not_smsa |  -.0167178   .0339236    -0.49   0.622    -.0832123    .0497767
             |
 c.age#c.age |  -.0012145   .0001968    -6.17   0.000    -.0016003   -.0008286
------------------------------------------------------------------------------
Underidentification test (Anderson canon. corr. LM statistic):          52.345
                                                   Chi-sq(2) P-val =    0.0000
------------------------------------------------------------------------------
Weak identification test (Cragg-Donald Wald F statistic):               21.286
Stock-Yogo weak ID test critical values: 10% maximal IV size             19.93
                                         15% maximal IV size             11.59
                                         20% maximal IV size              8.75
                                         25% maximal IV size              7.25
Source: Stock-Yogo (2005).  Reproduced by permission.
------------------------------------------------------------------------------
Sargan statistic (overidentification test of all instruments):           1.691
                                                   Chi-sq(1) P-val =    0.1935
------------------------------------------------------------------------------
Instrumented:         tenure
Included instruments: age not_smsa c.age#c.age
Excluded instruments: union south
Partialled-out:       _cons
                      nb: total SS, model F and R2s are after partialling-out;
                          any small-sample adjustments include partialled-out
                          variables in regressor count K
------------------------------------------------------------------------------

Absorbed degrees of freedom:
-----------------------------------------------------+
 Absorbed FE | Categories  - Redundant  = Num. Coefs |
-------------+---------------------------------------|
      idcode |      3461           0        3461     |
-----------------------------------------------------+

.