[回归分析求助] ivreg2 回归结果中只报告了 Hansen J statistic 的结果，求问工具变量是否有效？ [推广有奖]

3关注
5粉丝

已卖：1365份资源

博士生

33%

还不是VIP/贵宾

威望: 0 级
论坛币: 42482 个
通用积分: 14.2339
学术水平: 10 点
热心指数: 0 点
信用等级: 0 点
经验: 2636 点
帖子: 110
精华: 0
在线时间: 372 小时
注册时间: 2013-12-23
最后登录: 2025-12-18

楼主

听见海涛声 发表于 2018-4-30 23:27:03 |AI写论文

是否 +2 论坛币

k人参与回答

经管之家送您一份

应届毕业生专属福利!

求职就业群

赵安豆老师微信：zhaoandou666

经管之家联合CDA

送您一个全额奖学金名额~ !

立即领取

感谢您参与论坛问题回答

经管之家送您两个论坛币！

+2 论坛币

如题。
看到论坛很多学友使用ivreg2时，都会报告Weak identification test (Cragg-Donald Wald F statistic)等结果。

我使用ivreg2 回归结果中只报告了 Hansen J statistic 的结果。

求问如何知道回归中的工具变量是否有效？感谢各位大神指点！

代码如下：
ivreg2 roe (fds= l.fds lfn ) $z yr*, first r

回归结果如下：

First-stage regressions
-----------------------
Warning: collinearities detected among instruments
1st stage tests of excluded exogenous variables may be incorrect

First-stage regression of fds:

   Source |    SS          df    MS    Number of obs =    798
-------------+---------------------------------- F(16, 781)    = 1077.40
   Model |  2771.17492       16  173.198433 Prob > F       = 0.0000
Residual | 125.55047    781  .160756044 R-squared    = 0.9567
-------------+---------------------------------- Adj R-squared = 0.9558
   Total |  2896.72539    797  3.63453625 Root MSE       = .40094

------------------------------------------------------------------------------
      fds |    Coef. Std. Err.    t P>|t|    [95% Conf. Interval]
-------------+----------------------------------------------------------------
      ipr |  -.7784742 .123837 -6.29 0.000 -1.021567 -.5353815
      cash |  -.0937093 .1848735 -0.51 0.612 -.4566171 .2691985
      inta | .0729728 .3384978    0.22 0.829 -.5915005 .7374461
      inv | -.219052 .2984387 -0.73 0.463 -.8048889 .3667849
      lev | .088286 .0721323    1.22 0.221 -.0533102 .2298822
      size | .0150404 .0175326    0.86 0.391 -.0193762    .049457
      age |  -.0413863 .0294579 -1.40 0.160 -.0992123 .0164398
ownership |  -.0153685 .0380713 -0.40 0.687 -.0901028 .0593658
      yr1 |       0  (omitted)
      yr2 | .4988312 .0766778    6.51 0.000    .3483123 .6493502
      yr4 |  -.3122429 .0684479 -4.56 0.000 -.4466066 -.1778791
      yr5 |  -.0320953 .0657231 -0.49 0.625 -.1611101 .0969195
      yr6 |  -.1230321 .0644031 -1.91 0.056 -.2494557 .0033916
      yr7 | .6540823 .0651506 10.04 0.000    .5261913 .7819734
      yr8 | .401018 .0698746    5.74 0.000    .2638537 .5381824
         |
      fds |
      L1. | .9612215 .0097947 98.14 0.000    .9419944 .9804485
         |
      lfn | .1294921 .0210508    6.15 0.000    .0881692 .1708151
   _cons |  -.3833881 .3569935 -1.07 0.283 -1.084169 .3173924
------------------------------------------------------------------------------
Partial R-squared of excluded instruments: 0.9423
Test of excluded instruments:
  F(  2, 781) =  6377.56
  Prob > F    = 0.0000

Summary results for first-stage regressions:

            Shea
Variable    Partial R2    Partial R2    F(  2, 781) P-value
fds          0.9423       0.9423          6377.56       0.0000

IV (2SLS) regression with robust standard errors
------------------------------------------------

                                                   Number of obs =    798
                                                   F( 15, 781) =    9.56
                                                   Prob > F    = 0.0000
Total (centered) SS    =  6.939967727             Centered R2 = 0.2355
Total (uncentered) SS =  13.64917262             Uncentered R2 = 0.6113
Residual SS          =  5.305642002             Root MSE    =    .082

------------------------------------------------------------------------------
         |             Robust
      roe |    Coef. Std. Err.    z P>|z|    [95% Conf. Interval]
-------------+----------------------------------------------------------------
      fds | .0050063 .0015065    3.32 0.001    .0020537 .0079589
      ipr | .0697356 .024771    2.82 0.005    .0211854 .1182858
      cash | .3218028 .0404903    7.95 0.000    .2424433 .4011624
      inta |  -.3404941 .1222533 -2.79 0.005 -.5801062 -.1008821
      inv |  -.0265555 .067374 -0.39 0.693 -.1586062 .1054952
      lev |  -.0383824 .0291256 -1.32 0.188 -.0954676 .0187028
      size | .0298174 .0049024    6.08 0.000    .0202089 .0394258
      age |  -.0037206 .006417 -0.58 0.562 -.0162977 .0088565
ownership |  -.0165756 .0084315 -1.97 0.049    -.033101 -.0000501
      yr1 |       0  (omitted)
      yr2 |  -.0088435 .0231836 -0.38 0.703 -.0542825 .0365954
      yr4 |  -.0067761 .0169465 -0.40 0.689 -.0399906 .0264384
      yr5 |  -.0289826 .0163743 -1.77 0.077 -.0610756 .0031105
      yr6 |  -.0302858 .0151743 -2.00 0.046 -.0600268 -.0005448
      yr7 |  -.0355642 .0153144 -2.32 0.020 -.0655798 -.0055486
      yr8 |  -.0583836 .0167517 -3.49 0.000 -.0912162 -.0255509
   _cons |  -.5007741 .0970583 -5.16 0.000 -.6910049 -.3105432
------------------------------------------------------------------------------
Hansen J statistic (overidentification test of all instruments):       1.618
                                                Chi-sq(1) P-val = 0.20344
------------------------------------------------------------------------------
Collinearities detected among instruments: 1 instrument(s) dropped
Instrumented:  fds
Instruments: ipr cash inta inv lev size age ownership yr1 yr2 yr4 yr5 yr6 yr7
            yr8 L.fds lfn
------------------------------------------------------------------------------

.
end of do-file

.