Saving p value of z statistic after ivreg2

[Akanksha Aggarwal]

Hi.

For a falsification test, I want to run an IV regression and store the value of the coefficient of the instrumented variable and the corresponding p value for the coefficient.
I am able to store the coefficient value using the following:

matrix coef_IV = e(b)
scalar b = coef_IV[1,1]

What would be the command for saving the p value of the z statistic for the coefficient.

Thanks.

[Maarten Buis]

You can get the coefficient more easily: Say the intrumented variable is called x, then the coefficient is _b[x] . Simmilarly, the standard error is in _se[x]. When applicable, the degrees of freedom are stored in e(df_r). Now you have all the ingredients necessary for computing p-values and confidence intervals.

Alternatively you can look at r(table) which most estimation commands leave behind.

Here is an example. I showed the parts for pcturban, as with non-significant coefficients it is easier to see that computations are correct as there are more numbers to show instead of many 0s

. webuse hsng2, clear
(1980 Census housing data)

. ivregress 2sls rent pcturban (hsngval = faminc i.region), small


Instrumental variables 2SLS regression

      Source |       SS       df       MS         Number of obs   =         50
-------------+------------------------------      F(  2,    47)   =      42.66
       Model |  36677.4033     2  18338.7017      Prob > F        =     0.0000
    Residual |  24565.7167    47  522.674823      R-squared       =     0.5989
-------------+------------------------------      Adj R-squared   =     0.5818
       Total |    61243.12    49  1249.85959      Root MSE        =     22.862

------------------------------------------------------------------------------
        rent | Coefficient  Std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
     hsngval |   .0022398   .0003388     6.61   0.000     .0015583    .0029213
    pcturban |    .081516   .3081528     0.26   0.793    -.5384074    .7014394
       _cons |   120.7065   15.70688     7.68   0.000     89.10834    152.3047
------------------------------------------------------------------------------
Instrumented: hsngval
 Instruments: pcturban faminc 2.region 3.region 4.region

.
. di "p-value for pcturban = " 2*ttail(e(df_r),abs(_b[pcturban]/_se[pcturban]))
p-value for pcturban = .79252738

.
. di "95% ci of pcturban = [" _b[pcturban] - invt(e(df_r),0.975)*_se[pcturban] ///
>    " , " _b[pcturban] + invt(e(df_r),0.975)*_se[pcturban] "]"
95% ci of pcturban = [-.53840744 , .70143938]

.    
. matlist r(table)

             |   hsngval   pcturban      _cons
-------------+---------------------------------
           b |  .0022398    .081516   120.7065
          se |  .0003388   .3081528   15.70688
           t |  6.611874    .264531   7.684943
      pvalue |  3.17e-08   .7925274   7.55e-10
          ll |  .0015583  -.5384074   89.10834
          ul |  .0029213   .7014394   152.3047
          df |        47         47         47
        crit |  2.011741   2.011741   2.011741
       eform |         0          0          0

[Akanksha Aggarwal]

Thanks a lot.