Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
st: RE: Blown up IV coefficient
From
"Wooldridge, Jeffrey" <[email protected]>
To
<[email protected]>
Subject
st: RE: Blown up IV coefficient
Date
Wed, 9 Mar 2011 12:24:42 -0500
For the IV estimates to be at all convincing, first run the reduced
form,
areg roaddum rank agr_share ims_hab_pop, absorb(dcode) robust
and look at the sign of the coefficient on rank and its t statistic.
Hopefully you have a predicted sign for the effect of rank on roaddum.
If the sign is what you expect, is rank strongly statistically
significant? The statistics for weak identification provide this
information, too, but you should estimate the reduced form, anyway. It
is easier to see what is happening.
My guess is the instrument is weak. Or, the IV could actually be
endogenous (which cannot be tested, unfortunately).
Jeff
-----Original Message-----
From: [email protected]
[mailto:[email protected]] On Behalf Of Shikha Sinha
Sent: Wednesday, March 09, 2011 12:17 PM
To: [email protected]
Subject: st: Blown up IV coefficient
Dear all,
I am estimating an instrumental variable model by using -ivreg2 in
Stata. The IV estimates are way more than OLS estimate (4.006567).
Could you advise what could be possible reasons for such a blown up IV
coeff (49.97424 )? or my interpretation of coeff is incorrect.
areg agr_share roaddum ims_hab_pop, robust absorb(dcode)
Linear regression, absorbing indicators Number of obs =
266
F( 2, 250) =
0.66
Prob > F =
0.5182
R-squared =
0.0680
Adj R-squared =
0.0121
Root MSE =
26.294
------------------------------------------------------------------------
------
| Robust
agr_share | Coef. Std. Err. t P>|t| [95% Conf.
Interval]
-------------+----------------------------------------------------------
------
roaddum | 4.006567 3.572553 1.12 0.263 -3.02957
11.0427
ims_hab_pop | .001926 .0344155 0.06 0.955 -.0658552
.0697071
_cons | 34.78717 36.38659 0.96 0.340 -36.87615
106.4505
-------------+----------------------------------------------------------
------
dcode | absorbed (14
categories)
. xi: ivreg2 agr_share ims_hab_pop i.dcode (roaddum=rank), robust
i.dcode _Idcode_1-14 (_Idcode_1 for dcode==UP09
omitted)
IV (2SLS) estimation
--------------------
Estimates efficient for homoskedasticity only
Statistics robust to heteroskedasticity
Number of obs =
266
F( 15, 250) =
0.88
Prob > F =
0.5891
Total (centered) SS = 185457.6241 Centered R2 =
-0.5185
Total (uncentered) SS = 590824 Uncentered R2 =
0.5233
Residual SS = 281625.775 Root MSE =
32.54
------------------------------------------------------------------------
------
| Robust
agr_share | Coef. Std. Err. z P>|z| [95% Conf.
Interval]
-------------+----------------------------------------------------------
------
roaddum | 49.97424 35.09856 1.42 0.154 -18.81768
118.7662
ims_hab_pop | -.0354515 .0511597 -0.69 0.488 -.1357226
.0648196
_Idcode_2 | -39.0505 25.58464 -1.53 0.127 -89.19546
11.09446
_Idcode_3 | -41.62481 26.03165 -1.60 0.110 -92.64591
9.396298
_Idcode_4 | -16.54727 20.19038 -0.82 0.412 -56.11968
23.02513
_Idcode_5 | -41.45807 29.51111 -1.40 0.160 -99.29878
16.38264
_Idcode_6 | -32.93124 16.32568 -2.02 0.044 -64.929
-.9334909
_Idcode_7 | -8.880318 13.20529 -0.67 0.501 -34.76221
17.00157
_Idcode_8 | -44.38007 24.23395 -1.83 0.067 -91.87774
3.117595
_Idcode_9 | -44.81575 24.87717 -1.80 0.072 -93.57409
3.942605
_Idcode_10 | -21.14303 12.84912 -1.65 0.100 -46.32684
4.040777
_Idcode_11 | -22.56438 9.922245 -2.27 0.023 -42.01162
-3.117138
_Idcode_12 | -12.55903 12.52187 -1.00 0.316 -37.10144
11.98338
_Idcode_13 | -14.25003 9.233117 -1.54 0.123 -32.34661
3.846547
_Idcode_14 | -26.27491 11.42556 -2.30 0.021 -48.66859
-3.881229
_cons | 74.88709 53.28162 1.41 0.160 -29.54296
179.3171
------------------------------------------------------------------------
------
Underidentification test (Kleibergen-Paap rk LM statistic):
1.825
Chi-sq(1) P-val =
0.1767
------------------------------------------------------------------------
------
Weak identification test (Cragg-Donald Wald F statistic):
2.977
(Kleibergen-Paap rk Wald F statistic):
2.488
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: roaddum
Included instruments: ims_hab_pop _Idcode_2 _Idcode_3 _Idcode_4
_Idcode_5
_Idcode_6 _Idcode_7 _Idcode_8 _Idcode_9 _Idcode_10
_Idcode_11 _Idcode_12 _Idcode_13 _Idcode_14
Excluded instruments: rank
------------------------------------------------------------------------
------
Thanks,
Shikha
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/