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: Interpreting ivreg2 outputs
From
DE SOUZA Eric <[email protected]>
To
"[email protected]" <[email protected]>
Subject
st: RE: Interpreting ivreg2 outputs
Date
Sun, 17 Apr 2011 21:44:54 +0200
All the questions below are addressed either in the help file or in the following article which you can download for free:
http://www.stata-journal.com/article.html?article=st0030_3
Eric de Souza
College of Europe
Brugge (Bruges), Belgium
http://www.coleurope.eu
-----Original Message-----
From: [email protected] [mailto:[email protected]] On Behalf Of Hussain Samad
Sent: 17 April 2011 21:34
To: [email protected]
Subject: st: Interpreting ivreg2 outputs
Dear Statalist:
I'm trying to understand the output of ivreg2 command, especially various tests.
I very much appreciate if someone can explain. I have included necessary outputs
below. Please let me know if my understanding is correct or explain otherwise.
Endogeneity test:
P-val =0.0182, so exogeneity of the regressors is rejected at 5% level (that is, they are endogenous). Correct?
Overidentification test (Hansen J statistics):
P-val =0.1643, so overidentification restriction is satisfied (H0 cannot be rejected at 5% level). Correct?
Underidentification test (Kleibergen-Paap rk LM statistic):
P-val = 0.2358, so model is underidentified (H0 cannot be rejected at 5% level).
Correct? If so, what can be done to correct it?
Weak instrument test (Kleibergen-Paap rk Wald F statistic):
This is what I am most confused about. Which of the Stock-Yogo critical values I
compare with the F-stat? And what decision can be made based on my output? In case instruments are weak, what to do about it? Please guide.
Redundancy of instruments:
I know I can simply regress the endogenous variables on the instruments and check t-stats. But how can I come to a decision from the ivreg2 output? Or, what commands I should use after ivreg2?
I know there are lot of questions. But I do not know where else to go except for
here. Again, thanks a lot for your help.
-Hussain
Command:
ivreg2 lgpcexpfmv agehead sexhead educhead maxeducm maxeducf lgland lgnland lgvpop vwage* vpaved vprimschl vsecschl vmarket velec vgb vngo vsnet chardumm riverdum phighland1 rainfall (memprime mfonly=vprime*), endog(memprime mfonly)
cluster(vill) ffirst;
Outputs:
Summary results for first-stage regressions
-------------------------------------------
Variable | Shea Partial R2 | Partial R2 | F( 6, 290) P-value
memprime | 0.0382 | 0.0765 | 82.04
0.0000
mfonly | 0.0012 | 0.0024 | 1.58
0.1514
NB: first-stage F-stat cluster-robust
Underidentification tests
Ho: matrix of reduced form coefficients has rank=K1-1 (underidentified)
Ha: matrix has rank=K1 (identified)
Kleibergen-Paap rk LM statistic Chi-sq(5)=6.80 P-val=0.2358
Kleibergen-Paap rk Wald statistic Chi-sq(5)=8.42 P-val=0.1345
Weak identification test
Ho: equation is weakly identified
Kleibergen-Paap Wald rk F statistic 1.39
See main output for Cragg-Donald weak id test critical values
Weak-instrument-robust inference
Tests of joint significance of endogenous regressors B1 in main equation
Ho: B1=0 and overidentifying restrictions are valid
Anderson-Rubin Wald test F(6,290)= 1.60 P-val=0.1474
Anderson-Rubin Wald test Chi-sq(6)=9.68 P-val=0.1390
Stock-Wright LM S statistic Chi-sq(6)=7.32 P-val=0.2924
NB: Underidentification, weak identification and weak-identification-robust
test statistics cluster-robust
Number of clusters N_clust = 291
Number of observations N = 5207
Number of regressors K = 26
Number of instruments L = 30
Number of excluded instruments L1 = 6
IV (2SLS) estimation
--------------------
.
.
OUTPUT OMITTED ………………………
.
.
------------------------------------------------------------------------------
Underidentification test (Kleibergen-Paap rk LM statistic):
6.802
Chi-sq(5) P-val = 0.2358
------------------------------------------------------------------------------
Weak identification test (Kleibergen-Paap rk Wald F statistic):
1.391
Stock-Yogo weak ID test critical values: 5% maximal IV relative bias 15.72
10% maximal IV relative
bias 9.48
20% maximal IV relative bias
6.08
30% maximal IV relative bias
4.78
10% maximal IV size
21.68
15% maximal IV size
12.33
20% maximal IV size
9.10
25% maximal IV
size 7.42
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):
6.507
Chi-sq(4) P-val = 0.1643
-endog- option:
Endogeneity test of endogenous regressors:
8.013
Chi-sq(2) P-val = 0.0182
Regressors tested: memprime mfonly
------------------------------------------------------------------------------
Instrumented: memprime mfonly
Included instruments: agehead sexhead educhead maxeducm maxeducf lgland
lgnland lgvpop vwagem vwagef vwagec vpaved vprimschl
vsecschl vmarket velec vgb vngo vsnet chardumm
riverdum phighland1 rainfall Excluded instruments: vprime vprimeagehead vprimesexhead vprimeeduchead
vprimelgland vprimelgnland
------------------------------------------------------------------------------
*
* 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/