COWLES FOUNDATION FOR RESEARCH IN ECONOMICS
AT YALE UNIVERSITY

Box 208281
New Haven, CT 06520-8281

COWLES FOUNDATION DISCUSSION PAPER NO. 710RR

"Stability Comparisons of Estimators"

Donald W.K. Andrews

July 1984
Revised November 1985

This paper investigates a property of estimators called stability. The stability exponent of an estimator is defined to be a measure of the effect of any single observation in the sample on the realized value of the estimator. High stability is often desirable for robustness against misspecification and against highly variable observations.

Stability exponents are determined and compared for a wide variety of estimators and econometric models. They are found to depend on the maximal moment exponent (i.e., the number of finite moments) of the estimator’s influence curve. Since it is possible often to construct estimators with specified influence curves, estimators with different stability exponents can be construed.

See CFP 663