Abstract
The value for which the mean square error of a biased estimatoraT for the mean μ is less than the variance of an unbiased estimatorT is derived by minimizingMSE(aT). The resulting optimal value is 1/[1+c(n)v 2], wherev=σ/μ, is the coefficient of variation. WhenT is the UMVUE\(\bar X\), thenc(n)=1/n, and the optimal value becomes 1/(n+v 2) (Searls, 1964). Whenever prior information about the size ofv is available the shrinkage procedure is useful. In fact for some members of the one-parameter exponential families it is known that the variance is at most a quadratic function of the mean. If we identify the pertinent coefficients in the quadratic function, it becomes easy to determinev.
Similar content being viewed by others
References
Arnholt AT, Hebert JL (1995) Estimating the mean with known coefficient of variation.The American Statistician,49, 367–369.
Bibby J (1972) Minimum mean square error estimation, ridge regression and some unanswered questions.Progress in Statistics 1, 107–121.
Bibby J, Toutenburg H (1977)Prediction and Improved Estimation in Linear Models. Wiley, Chichester.
Bibby J, Toutenburg H (1978) Improved estimation and prediction.Zeitschrift für angewandte Mathematik und Mechanik 58, 45–49.
Bickel JP, Doksum KA (2000)Mathematical Statistics, Basic Ideas and Selected Topics, Vol. 1 (2nd Ed.). Prentice-Hall, New Jersey.
Gleser LJ, Healy JD (1976) Estimating the mean of a normal distribution with known coefficient of variation.Journal of the American Statistical Association 71, 977–981.
Khan RA (1968) A note on estimating the mean of a normal distribution with known coefficient of variation.Journal of the American Statistical Association 63, 1039–1041.
Kleffe J (1985) Some remarks on improving unbiased estimators by multiplication with a constant.Linear Statistical Inference. Ed.: Calinski, T. and Klonecki, W., 150–161. Cambridge, Massachussets.
Morris CN (1982) Natural exponential families with quadratic variance functions.The Annals of Statistics 10, 65–80.
Searls DT (1964) The utilization of a known coefficient of variation in the estimation procedure.Journal of the American Statistical Association 59, 1225–1226.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Wencheko, E., Wijekoon, P. Improved estimation of the mean in one-parameter exponential families with known coefficient of variation. Statistical Papers 46, 101–115 (2005). https://doi.org/10.1007/BF02762037
Received:
Revised:
Issue Date:
DOI: https://doi.org/10.1007/BF02762037