The problem of instrument proliferation and its consequences—overfitting of the endogenous explanatory variables, biased instrumental-variables and generalized method of moments estimators, and weakening of the power of the overidentification tests—are well known. This article introduces a statistical method to reduce the instrument count. Principal component analysis is applied on the instrument matrix, and the principal-component analysis scores are used as instruments for the panel generalized method of moments estimation. This strategy is implemented through the new command pca2.
AltonjiJ. G., and SegalL. M.1996. Small-sample bias in GMM estimation of covariance structures. Journal of Business and Economic Statistics14: 353–366.
2.
AmemiyaT.1966. On the use of principal components of independent variables in two-stage least-squares estimation. International Economic Review7: 283–303.
3.
AndersonT. G., and SørensonB. E.1996. GMM estimation of a stochastic volatility model: A Monte Carlo study. Journal of Business and Economic Statistics14: 328–352.
4.
AndrewsD. W. K.1999. Consistent moment selection procedures for generalized method of moments estimation. Econometrica67: 543–564.
5.
AndrewsD. W. K., and LuB.2001. Consistent model and moment selection procedures for GMM estimation with application to dynamic panel data models. Journal of Econometrics101: 123–164.
6.
ArellanoM., and BondS.1991. Some tests of specification for panel data: Monte Carlo evidence and an application to employment equations. Review of Economic Studies58: 277–297.
7.
ArellanoM., and BoverO.1995. Another look at the instrumental variable estimation of error-components models. Journal of Econometrics68: 29–51.
8.
BaiJ., and NgS.2008. Forecasting economic time series using targeted predictors. Journal of Econometrics146: 304–317.
9.
BaiJ., and NgS.2010. Instrumental variable estimation in a data rich environment. Econometric Theory26: 1577–1606.
10.
BekkerP. A.1994. Alternative approximations to the distributions of instrumental variable estimators. Econometrica62: 657–681.
11.
BelloniA., ChenD., ChernozhukovV., and HansenC.2012. Sparse models and methods for optimal instruments with an application to eminent domain. Econometrica80: 2369–2429.
12.
BlundellR., and BondS.1998. Initial conditions and moment restrictions in dynamic panel data models. Journal of Econometrics87: 115–143.
13.
BoundJ., JaegerD. A., and BakerR. M.1995. Problems with instrumental variables estimation when the correlation between the instruments and the endogeneous explanatory variable is weak. Journal of the American Statistical Association90: 443–450.
14.
BowsherC. G.2002. On testing overidentifying restrictions in dynamic panel data models. Economics Letters77: 211–220.
DagenaisM. G., and DagenaisD. L.1997. Higher moment estimators for linear regression models with errors in the variables. Journal of Econometrics76: 193–221.
18.
DoranH. E., and SchmidtP.2006. GMM estimators with improved finite sample properties using principal components of the weighting matrix, with an application to the dynamic panel data model. Journal of Econometrics133: 387–409.
19.
GolinelliR., and MomiglianoS.2009. The cyclical reaction of fiscal policies in the Euro area: The role of modelling choices and data vintages. Fiscal Studies30: 39–72.
20.
GroenJ. J. J., and KapetaniosG.2009. Parsimonious estimation with many instruments. Staff Report 386, Federal Reserve Bank of New York.
21.
HanC., and PhillipsP. C. B.2006. GMM with many moment conditions. Econometrica74: 147–192.
22.
KapetaniosG., and MarcellinoM.2010. Factor-GMM estimation with large sets of possibly weak instruments. Computational Statistics and Data Analysis54: 2655–2675.
23.
KivietJ. F.1995. On bias, inconsistency, and efficiency of various estimators in dynamic panel data models. Journal of Econometrics68: 53–78.
24.
KloekT., and MennesL. B. M.1960. Simultaneous equations estimation based on principal components of predetermined variables. Econometrica28: 45–61.
25.
LiaoZ.2013. Adaptive GMM shrinkage estimation with consistent moment selection. Econometric Theory29: 857–904.
26.
ManderA.2006. lars: Stata module to perform least angle regression. Statistical Software Components S456860, Department of Economics, Boston College. https://ideas.repec.org/c/boc/bocode/s456860.html.
27.
MehrhoffJ.2009. A solution to the problem of too many instruments in dynamic panel data GMM. Discussion Paper No. 1/2009, Deutsche Bundesbank.
28.
NeweyW. K., and SmithR. J.2004. Higher order properties of GMM and generalized empirical likelihood estimators. Econometrica72: 219–255.
29.
RoodmanD.2009a. A note on the theme of too many instruments. Oxford Bulletin of Economics and Statistics71: 135–158.
30.
RoodmanD.2009b. How to do xtabond2: An introduction to difference and system GMM in Stata. Stata Journal9: 86–136.
31.
SarganJ. D.1958. The estimation of economic relationships using instrumental variables. Econometrica26: 393–415.
32.
StaigerD., and StockJ. H.1997. Instrumental variables regression with weak instruments. Econometrica65: 557–586.
33.
StockJ. H., and WatsonM. W.2002. Forecasting using principal components from a large number of predictors. Journal of the American Statistical Association97: 1167–1179.
34.
ZiliakJ. P.1997. Efficient estimation with panel data when instruments are predetermined: An empirical comparison of moment-condition estimators. Journal of Business and Economic Statistics14: 419–431.
