Jackknife Instrumental Variables Estimation

Two-stage-least-squares (2SLS) estimates are biased towards OLS estimates. This bias grows with the degree of over-identification and can generate highly misleading results. In this paper we propose two simple alternatives to 2SLS and limited-information-maximum-likelihood (LIML) estimators for models with more instruments than endogenous regressors. These estimators can be interpreted as instrumental variables procedures using an instrument that is independent of disturbances even in finite samples. Independence is achieved by using a `leave-one-out' jackknife-type fitted value in place of the usual first-stage equation. The new estimators are first-order equivalent to 2SLS but with finite-sample properties superior to those of 2SLS and similar to LIML when there are many instruments. Moreover, the jackknife estimators appear to be less sensitive than LIML to deviations from the linear reduced form used in classical simultaneous equations models.

[1]  Thomas M. Stoker.  Smoothing bias in the measurement of marginal effects , 1996 .

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

[3]  Jinook Jeong,et al.  ON THE EXACT SMALL SAMPLE DISTRIBUTION OF THE INSTRUMENTAL VARIABLE ESTIMATOR , 1992 .

[4]  Joshua D. Angrist,et al.  Lifetime Earnings and the Vietnam Era Draft Lottery: Evidence from Social Security Administrative Records , 1990 .

[5]  F. Fisher.  Approximate Specification and the Choice of a k-Class Estimator , 1967 .

[6]  R. Cook.  Influential Observations in Linear Regression , 1979 .

[7]  T. Sawa.  Almost Unbiased Estimator in Simultaneous Equations Systems , 1973 .

[8]  Gary Chamberlain,et al.  Hierarchical Bayes Models with Many Instrumental Variables , 1996 .

[9]  F. Fisher.  The Relative Sensitivity to Specification Error of Different k-Class Estimators , 1966 .

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

[11]  Paul A. Bekker,et al.  ALTERNATIVE APPROXIMATIONS TO THE DISTRIBUTIONS OF INSTRUMENTAL VARIABLE ESTIMATORS , 1994 .

[12]  Peter C. B. Phillips,et al.  Exact Small Sample Theory in the Simultaneous Equations Model , 1983 .

[13]  J. Angrist,et al.  Does Compulsory School Attendance Affect Schooling and Earnings? , 1990 .

[14]  A. Buse,et al.  The Bias of Instrumental Variable Estimators , 1992 .

[15]  J. Angrist,et al.  The Effect of Age at School Entry on Educational Attainment: An Application of Instrumental Variables with Moments from Two Samples , 1990 .

[16]  J. Angrist,et al.  Split Sample Instrumental Variables , 1995 .

[17]  C. Nelson,et al.  The Distribution of the Instrumental Variables Estimator and its T-Ratiowhen the Instrument is a Poor One , 1988 .

[18]  H. James.  VARIETIES OF SELECTION BIAS , 1990 .

[19]  Joshua D. Angrist,et al.  Split-Sample Instrumental Variables Estimates of the Return to Schooling , 1995 .

[20]  A. L. Nagar.  The Bias and Moment Matrix of the General k-Class Estimators of the Parameters in Simultaneous Equations , 1959 .