Normal maximum likelihood, weighted least squares, and ridge regression estimates

• Autorzy: Withers C. S. , Nadarajah S.
• Języki publikacji: EN

Abstrakty

EN

There have been many papers published (in almost every statistics related journal) suggesting that normal maximum likelihood is superior or inferior to weighted least squares and other approaches. In this note, we show that the three main estimation methods (normal maximum likelihood, weighted least squares and ridge regression) all have the same asymptotic covariance and that there is no gain in efficiency among them. We also show how the bias of these estimators can be reduced and conduct a simulation study to illustrate the magnitude of bias reduction.

Słowa kluczowe

EN

bias reduction; normal maximum likelihood; regression; weighted least squares

• Wydawca: Wrocław University of Science and Technology
• Czasopismo: Probability and Mathematical Statistics
• Rocznik: 2012
• Tom: Vol. 32, Fasc. 1
• Strony: 11--24
• Opis fizyczny: Bibliogr. 13 poz., wykr.

Twórcy

autor

Withers C. S.

• Applied Mathematics Group, Industrial Research Limited, Lower Hutt, New Zealand

autor

Nadarajah S.

• School of Mathematics, University of Manchester, Manchester M13 9PL, United Kingdom

Bibliografia

• [1] H. Abdi, Least squares, in: Encyclopedia of Social Sciences Research Methods, M. Lewis-Beck, A. Bryman and T. Futing (Eds.), Sage, Thousand Oaks, California, 2003.
• [2] A. Azzalini, A class of distributions include the normal ones, Scand. J. Statist. 12 (1985), pp. 171-178.
• [3] J. Behboodian, A. Jamalizadeh and N. Balakrishnan, A new class of skew-Cauchy distributions, Statist. Probab. Lett. 76 (2006), pp. 1488-1493.
• [4] J.-M. Bonny, M. Zanca, J.-Y. Boire and A. Veyre, T2 maximum likelihood estimation from multiple spin-echo magnitude images, Magnet. Reson. Med. 36 (1996), pp. 287-293.
• [5] S. G. Candy, A. J. Constable, T. Lamb and R. Williams, A Von Bertalanffy growth model for toothfish at Heard Island fitted to length-at-age data and compared to observed growth from mark-recapture studies, CCAMLR Science 14 (2007), pp. 43-66.
• [6] K. Emrich and W. Urfer, Benefits and complications of maximum likelihood estimation in (composite) interval mapping methods using EM and ECM, Euphytica 137 (2004), pp. 155-163.
• [7] J. A. Hausman and D. A. Wise, Stratification on endogenous variables and estimation: the Gary income maintenance experiment, in: Structural Analysis of Discrete Data with Econometric Application, C. Manski and D. McFadden (Eds.), Massachusetts Institute of Technology Press, Cambridge, Massachusetts, 1981, pp. 364-391.
• [8] J. Jiang, Linear and Generalized Linear Mixed Models and Their Applications, Springer, New York 2007.
• [9] K. G. Mehrotra, P. M. Kulkarni, R. M. Tripathi and J. E. Michalek, Maximum likelihood estimation for longitudinal data with truncated observations, Stat. Med. 19 (2000), pp. 2975-2988.
• [10] M. M. Olsen, J. Swevers and W. Verdonck, Maximum likelihood identification of a dynamic robot model: implementation issues, Int. J. Robot. Res. 21 (2002), pp. 89-96.
• [11] C. S. Withers, Expansions for the distribution and quantiles of a regular functional of the empirical distribution with applications to nonparameteric confidence intervals, Ann. Statist. 11 (1983), pp. 577-587.
• [12] C. S. Withers, Bias reduction by Taylor series, Commun. Stat. - Theory and Methods 16 (1987), pp. 2369-2384.
• [13] C. S. Withers, Nonparametric confidence intervals for functions of several distributions, Ann. Inst. Statist. Math. 40 (1988), 727-746.