Helmut Farbmacher, "GMM with Many Weak Moment Conditions: Replication 
and Application of Newey and Windmeijer (2009)", Journal of Applied 
Econometrics, Vol. 27, No. 2, 2012, pp. 343-346.

The file farbmacher.zip contains 13 Stata files: six do files, six ado 
files, and a help file. These are all ASCII files in DOS format. 
Unix/Linux users should use "unzip -a".

------

1) ivreg3.ado

Estimates the CUE with usual standard errors and with the many weak
instruments standard errors of Newey and Windmeijer (type "help ivreg3" in
Stata to call ivreg3.sthlp).

------

Table 1: Replication of Newey and Windmeijer (2009):

2) MCsim_2sls_jive2.do

Runs the Monte Carlo simulations for 2SLS and JIVE2. It calls
"mc_ivreg2_2sls.ado" and "mc_jive2.ado".

2a) mc_ivreg2_2sls.ado

Contains the definition of the model and runs the IV regression.

2b) mc_jive2.ado

Contains the definition of the model and estimates JIVE2. Stata's jive
command can be downloaded from
http://www.stata-journal.com/software/sj6-3/st0108/.

3) MCsim_ivreg3.do

Runs the Monte Carlo simulations for CUEC. It calls "mc_ivreg3_sim.ado". The
CUE is estimated with different starting values of beta (-2,-1,0,1,2). The
used estimates are obtained from the best model. 

3a) mc_ivreg3_sim.ado

Defines the model and estimates the CUE with NW's variance estimator using
"ivreg3.ado".  

------

Table 2: Application of NW's new variance estimator to the dataset from
Angrist and Krueger(1991):
http://econ-www.mit.edu/faculty/angrist/data1/data/angkru1991

4) QoB Table V.do

Draws 100 random subsamples and estimates the effect of schooling with 2SLS,
JIVE2, CUE and CUEC.

------

Figure 1: 

5) graph_rho.do, graph_cp.do and graph_ivs.do generate the datasets for
figure 1. They call mc_ivreg2_2sls.ado, mc_jive2.ado, mc_ivreg3_nwind.ado
and mc_ivreg3.ado. 

5a) mc_ivreg3_nwind.ado 

Defines the model and estimates the CUE with NW's variance estimator using
"ivreg3.ado".

5b) mc_ivreg3.ado

Defines the model and estimates the CUE with usual standard errors using
"ivreg3.ado".

------

References:

Newey W.K., Windmeijer F. 2009. Generalized method of moments with many weak
moment conditions. Econometrica 77(3): 687-719.

Angrist J., Krueger A. 1991. Does compulsory school attendance affect
schooling and earnings. Quarterly Journal of Economics 106(4): 979-1014. 

------

Please address any questions to:

Helmut Farbmacher
Munich Center for the Economics of Aging (MEA)
Max Planck Society
Amalienstr. 33
80799 Munich
Germany

Email: farbmacher [AT] mea.mpisoc.mpg.de