Abstract
We consider the problem of estimating the scale parameter θ of the shifted exponential distribution with unknown shift based on a set of observed records drawn from a sequential sample of independent and identically distributed random variables. Under a large class of bowl-shaped loss functions, the best affine equivariant estimator (BAEE) of θ is shown to be inadmissible. Two dominating procedures are proposed. A numerical study is performed to show the extent of risk reduction that the improved estimators provide over the BAEE.
Similar content being viewed by others
References
Ahsanullah M (1995) Introduction to record statistics. NOVA Science, Huntington New York USA
Ahsanullah M, Nevzorov VB (2001) Ordered random variables. NOVA Science, Huntington New York USA
Arnold BC, Balakrishnan N, Nagaraja HN (1998) Records. Wiley, New York
Brewster JF (1974) Alternative estimators for the scale parameter of the exponential distribution with unknown location. Ann Stat 2:553–557
Brewster JF, Zidek JV (1974) Improving on equivariant estimators. Ann Stat 2:21–38
Elfessi A (1997) Estimation of a linear function of the parameters of an exponential distribution from doubly censored samples. Stat Prob Lett 36:251–259
Madi MT, Tsui KW (1990) Estimation of the ratio of the scale parameters of two exponential distributions with unknown location parameters. Ann Inst Stat Math 42:77–87
Madi MT (1993) Alternative estimators for the variance of several normal distributions. Stat Proba Lett 17:321–328
Madi MT (2002) On the invariant estimation of an exponential scale using doubly censored data. Stat Proba Lett 56:77–82
Rukhin AL, Zidek JV (1985) Estimation of linear parametric functions for several exponential samples. Stat Decis 3:225–238
Zidek JV (1973) Estimating the scale parameter of the exponential distribution with unknown locations. Ann Stat 1:264–278
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Madi, M.T. Decision theoretic estimation using record statistics. Metrika 63, 91–97 (2006). https://doi.org/10.1007/s00184-005-0009-8
Received:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00184-005-0009-8