TY  - JOUR
AU  - Angrist,Joshua D.
AU  - Imbens,Guido W.
AU  - Krueger,Alan
TI  - Jackknife Instrumental Variables Estimation
JF  - National Bureau of Economic Research Technical Working Paper Series
VL  - No. 172
PY  - 1995
Y2  - February 1995
UR  - http://www.nber.org/papers/t0172
L1  - http://www.nber.org/papers/t0172.pdf
N1  - Author contact info:
Joshua Angrist
Department of Economics
MIT, E52-353
50 Memorial Drive
Cambridge, MA  02142-1347
Tel: 617/253-8909
Fax: 617/253-1330
E-Mail: angrist@mit.edu

Guido Imbens
Department of Economics
Littauer Center
Harvard University
1805 Cambridge Street
Cambridge, MA 02138
Tel: 617/384-7485
Fax: 617/495-7730
E-Mail: imbens@fas.harvard.edu

Alan B. Krueger
Industrial Relations Section
Firestone Library
Princeton University
Princeton, NJ 08544
Tel: 609/258-4046
Fax: 609/258-2907
E-Mail: akrueger@princeton.edu
AB  - 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.
ER  -