Remote Access Proceedings of the American Mathematical Society
Green Open Access

Proceedings of the American Mathematical Society

ISSN 1088-6826(online) ISSN 0002-9939(print)



Wald's equation and asymptotic bias
of randomly stopped $U$-statistics

Authors: Victor H. de la Peña and Tze Leung Lai
Journal: Proc. Amer. Math. Soc. 125 (1997), 917-925
MSC (1991): Primary 60G40, 62L12; Secondary 62L10
MathSciNet review: 1350937
Full-text PDF

Abstract | References | Similar Articles | Additional Information

Abstract: In this paper we make use of decoupling arguments and martingale inequalities to extend Wald's equation for sample sums to randomly stopped de-normalized $U$-statistics. We also apply this result in conjunction with nonlinear renewal theory to obtain asymptotic expansions for the means of normalized $U$-statistics from sequential samples.

References [Enhancements On Off] (What's this?)

  • [AW] Aras, G. and Woodroofe, M., Asymptotic expansions for the moments of a randomly stopped average, Ann. Statist. 21 (1993), 503-519. MR 94a:62120
  • [CdlPT] Chow, Y. S., de la Pe$\tilde \mathrm {n}$a, V. H. and Teicher, H., Wald's equation for a class of de-normalized $U$-statistics, Ann. Probab. 21 (1993), 1151-1158. MR 94c:60064
  • [CT] Chow, Y. S., and Teicher, H., Probability Theory: Independence, Interchangeability, Martingales, Springer-Verlag, New York, 1978. MR 80a:60004
  • [Co] Cox, D. R., A note on sequential estimation of means, Proc. Cambridge Philos. Soc. 48 (1952), 447-450. MR 14:190e
  • [GS] Grams, W. F. and Serfling, R. J., Convergence rates for $U$-statistics and related statistics, Ann. Statist. 1 (1973), 153-160. MR 49:1561
  • [Ha] Halmos, P. R., The theory of unbiased estimation, Ann. Math. Statist. 17 (1946), 34-43. MR 7:463g
  • [H] Hoeffding, W., A class of statistics with asymptotically normal distribution, Ann. Math. Statist. 19 (1948), 293-325. MR 10:134g
  • [K] Klass, M. J., A method of approximating expectations of functions of sums of independent random variables, Ann. Probab. 9 (1981), 413-428. MR 82f:60119
  • [LS] Lai, T. L. and Siegmund, D., A nonlinear renewal theory with applications to sequential analysis I, Ann. Statist. 5 (1977), 946-954; II, Ann. Statist. 7 (1979), 60-76. MR 56:3935; MR 80c:62100
  • [LW] Lai, T. L. and Wang, J. Q., Edgeworth expansions for symmetric statistics with applications to bootstrap methods, Statistica Sinica 3 (1993), 517-542. MR 95c:62019
  • [S1] Siegmund, D., Estimation following sequential tests, Biometrika 65 (1978), 341-349. MR 80f:62079
  • [S2] Siegmund, D., Sequential Analysis: Tests and Confidence Intervals, Springer-Verlag, New York, 1985. MR 87h:62145
  • [W1] Woodroofe, M., Second order approximation for sequential point and interval estimation, Ann. Statist. 5 (1977), 985-995. MR 58:13548
  • [W2] Woodroofe, M., Nonlinear Renewal Theory in Sequential Analysis, SIAM, Philadelphia, 1982. MR 83j:62118
  • [Z] Zinn, J., Comparison of martingale difference sequences, Probability in Banach spaces V, Lecture Notes in Math., vol. 1153, Springer-Verlag, Berlin, 1985, pp. 453-457. MR 87f:60070

Similar Articles

Retrieve articles in Proceedings of the American Mathematical Society with MSC (1991): 60G40, 62L12, 62L10

Retrieve articles in all journals with MSC (1991): 60G40, 62L12, 62L10

Additional Information

Victor H. de la Peña
Affiliation: Department of Statistics, Columbia University, 617 Mathematics Bldg., New York, New York 10027

Tze Leung Lai
Affiliation: Department of Statistics, Stanford University, Sequoia Hall, Stanford, California 94305-4065

Keywords: Hoeffding decomposition, decoupling, martingales, Wald's equation, stopping times
Received by editor(s): October 15, 1994
Received by editor(s) in revised form: July 28, 1995
Additional Notes: The first author’s research was supported by the National Science Foundation under DMS-9310682.
The second author’s research was supported by the National Science Foundation under DMS-9403794.
Communicated by: Wei Y. Loh
Article copyright: © Copyright 1997 American Mathematical Society

American Mathematical Society