A Note on the Theme of Too Many Instruments

The difference and system generalized method of moments (GMM) estimators are growing in popularity. As implemented in popular software, the estimators easily generate instruments that are numerous and, in system GMM, potentially suspect. A large instrument collection overfits endogenous variables even as it weakens the Hansen test of the instruments’ joint validity. This paper reviews the evidence on the effects of instrument proliferation, and describes and simulates simple ways to control it. It illustrates the dangers by replicating Forbes [American Economic Review (2000) Vol. 90, pp. 869–887] on income inequality and Levine et al. [Journal of Monetary Economics] (2000) Vol. 46, pp. 31–77] on financial sector development. Results in both papers appear driven by previously undetected endogeneity.

[1]  Stan Hurn.  Panel Data Econometrics , 2010 .

[2]  Gordon Tullock.  Publication Decisions and Tests of Significance—A Comment , 1959 .

[3]  M. Arellano,et al.  Some Tests of Specification for Panel Data: Monte Carlo Evidence and an Application to Employment Equations , 1991 .

[4]  Kristin J. Forbes,et al.  A Reassessment of the Relationship between Inequality and Growth , 2000 .

[5]  Jurgen A. Doornik,et al.  Panel Data estimation using DPD for Ox , 2001 .

[6]  Joseph G. Altonji,et al.  Small Sample Bias in GMM Estimation of Covariance Structures , 1994 .

[7]  R. Blundell,et al.  Initial Conditions and Moment Restrictions in Dynamic Panel Data Models , 1998 .

[8]  T. Sterling.  Publication Decisions and their Possible Effects on Inferences Drawn from Tests of Significance—or Vice Versa , 1959 .

[9]  Stephen Bond,et al.  Dynamic panel data models: a guide to micro data methods and practice , 2002 .

[10]  T. Beck,et al.  Financial Intermediation and Growth: Causality and Causes , 1999 .

[11]  M. Carkovic,et al.  Does Foreign Direct Investment Accelerate Economic Growth? , 2002 .

[12]  Norman V. Loayza,et al.  Determinants of Current Account Deficits in Developing Countries , 1999 .

[13]  David Roodman.  How to do Xtabond2: An Introduction to Difference and System GMM in Stata , 2006 .

[14]  Anke Hoeffler,et al.  GMM Estimation of Empirical Growth Models , 2001 .

[15]  W. Newey,et al.  Estimating vector autoregressions with panel data , 1988 .

[16]  Jeffrey M. Woodbridge.  Econometric Analysis of Cross Section and Panel Data , 2002 .

[17]  Cheng Hsiao,et al.  Formulation and estimation of dynamic models using panel data , 1982 .

[18]  Klaus Deininger,et al.  A New Data Set Measuring Income Inequality , 1996 .

[19]  J. Stock,et al.  Instrumental Variables Regression with Weak Instruments , 1994 .

[20]  James P. Ziliak.  Efficient Estimation with Panel Data When Instruments Are Predetermined: An Empirical Comparison of Moment-Condition Estimators , 1997 .

[21]  J. Sargan.  THE ESTIMATION OF ECONOMIC RELATIONSHIPS USING INSTRUMENTAL VARIABLES , 1958 .

[22]  L. Bauwens,et al.  Econometrics , 2005 .

[23]  Karl Rihaczek,et al.  1. WHAT IS DATA MINING? , 2019, Data Mining for the Social Sciences.

[24]  George Tauchen,et al.  Statistical Properties of Generalized Method-of-Moments Estimators of Structural Parameters Obtained From Financial Market Data , 1986 .

[25]  M. Arellano,et al.  Another look at the instrumental variable estimation of error-components models , 1995 .

[26]  Stephen Bond,et al.  GMM Estimation with persistent panel data: an application to production functions , 1999 .

[27]  F. Windmeijer.  A Finite Sample Correction for the Variance of Linear Two-Step GMM Estimators , 2000 .

[28]  Ian Witten,et al.  Data Mining , 2000 .

[29]  Stephen Bond,et al.  Estimation in dynamic panel data models: Improving on the performance of the standard GMM estimator , 2000 .

[30]  L. Hansen.  Large Sample Properties of Generalized Method of Moments Estimators , 1982 .

[31]  E. Feige.  The Consequences of Journal Editorial Policies and a Suggestion for Revision , 1975, Journal of Political Economy.

[32]  Tommy Stanley,et al.  Meta-Regression Methods for Detecting and Estimating Empirical Effects in the Presence of Publication Selection , 2007 .

[33]  Bent E. Sørensen,et al.  GMM Estimation of a Stochastic Volatility Model: A Monte Carlo Study , 1996 .

[34]  M. Nerlove,et al.  Biases in dynamic models with fixed effects , 1988 .

[35]  Clive G. Bowsher.  On testing overidentifying restrictions in dynamic panel data models , 2002 .