zbMATH — the first resource for mathematics

Asymptotic properties of a stochastic EM algorithm for mixtures with censored data. (English) Zbl 1178.62020
Summary: Weak consistency and asymptotic normality are shown for a stochastic EM algorithm for censored data from a mixture of distributions under lognormal assumptions. The asymptotic properties hold for all parameters of the distributions, including the mixing parameter. In order to make parameter estimation meaningful it is necessary to know that the censored mixture distribution is identifiable. General conditions under which this is the case are given. The stochastic EM algorithm addressed in this paper is used for estimation of wood fibre length distributions based on optically measured data from cylindric wood samples (increment cores).

62F12 Asymptotic properties of parametric estimators
65C60 Computational problems in statistics (MSC2010)
62N01 Censored data models
62F10 Point estimation
62P10 Applications of statistics to biology and medical sciences; meta analysis
Full Text: DOI
[1] Ahmad, E.K., Identifiability of finite mixtures using a new transform, Ann. inst. statist. math., 40, 261-265, (1988) · Zbl 0668.62011
[2] Atienza, N.; Garcia-Heras, J.; Munoz-Pichardo, J.M., A new condition for identifiability of finite mixture distributions, Metrika, 63, 215-221, (2006) · Zbl 1095.62016
[3] Bergsten, U.; Lindeberg, J.; Rindby, A.; Evans, R., Batch measurements of wood density on intact radial increment cores using X-ray densitometry, Wood sci. technol., 35, 435-452, (2001)
[4] Biscarat, J.C., Almost sure convergence of a class of stochastic algorithms, Stochastic process. appl., 50, 83-99, (1994) · Zbl 0819.60033
[5] Caffo, B.S.; Jank, W.; Jones, G.L., Ascent-based Monte Carlo expectation-maximization, J. roy. statist. soc. ser. B, 67, 235-251, (2005) · Zbl 1075.65011
[6] Casella, G.; Berger, R., Statistical inference, (1990), Duxbury Press Belmont, CA · Zbl 0699.62001
[7] Chan, K.S.; Ledolter, J., Monte Carlo EM estimation for time series models involving counts, J. amer. statist. assoc., 90, 242-252, (1995) · Zbl 0819.62069
[8] Chauveau, D., A stochastic EM algorithm for mixtures with censored data, J. statist. plann. inference, 46, 1-25, (1995) · Zbl 0821.62013
[9] Dempster, A.P.; Laird, N.M.; Rubin, D.B., Maximum likelihood from incomplete data via the EM algorithm (with discussion), J. roy. statist. soc. ser. B, 39, 1-38, (1977) · Zbl 0364.62022
[10] Downes, G.M., Hudson, I.L., Raymond, C.A., Michell, A.M., Schimleck, L.S., Evans, R., Dean G.H., 1997. Sampling Plantation Eucalypts for Wood and Fibre Properties. CSIRO Publishing, Melbourne, 132pp.
[11] Evans, R.; Downes, G.; Menz, D.; Stringer, S., Rapid measurement of variation in tracheid transverse dimensions in a radiata pine tree, Appita J., 48, 134-138, (1995)
[12] Evans, R.; Ilic, J., Rapid prediction of wood stiffness from microfibril angle and density, Forest prod. J., 51, 53-57, (2001)
[13] Everitt, B.S.; Hand, D.J., Finite mixture distributions, (1981), Chapman & Hall London · Zbl 0466.62018
[14] Franklin, G.L., Preparation of thin sections of synthetic resins and wood resin composites and a new macerating method for wood, Nature, 155, 51, (1945)
[15] Ilvessalo-Pfäffli, M.-S., Fiber atlas—identification of papermaking fibers, (1995), Springer Berlin
[16] Jank, W., Implementing and diagnosing the stochastic approximation EM algorithm, J. comput. graph. statist., 15, 803-829, (2006)
[17] Krantz, S.G.; Parks, H.R., A primer of real analytic functions, (1992), Birkhäuser Basel · Zbl 0767.26001
[18] Laslett, G.M., The survival curve under monotone density constraints with applications to two-dimensional line segment processes, Biometrika, 69, 153-160, (1982)
[19] Mörling, T.; Sjöstedt-deLuna, S.; Svensson, I.; Fries, A.; Ericsson, T., A method to estimate fibre length distribution in conifers based on wood samples from increment cores, Holzforschung, 57, 248-254, (2003)
[20] Nielsen, S.F., The stochastic EM algorithm: estimation and asymptotic results, Bernoulli, 6, 457-489, (2000) · Zbl 0981.62022
[21] Ranneby, B.; Cruse, T.; Hägglund, B.; Jonasson, H.; Swärd, J., Designing a new national forest survey for Sweden, Studia forestalia suecica, 177, 1-29, (1987)
[22] Redner, R.A.; Walker, H.F., Mixture densities, maximum likelihood and the EM algorithm, SIAM rev., 26, 195-239, (1984) · Zbl 0536.62021
[23] Sherman, R.P.; Ho, Y.K.; Dalal, S.R., Conditions for convergence of Monte Carlo EM sequences with an application to product diffusion modeling, Econom. J., 2, 248-267, (1999) · Zbl 0955.91050
[24] Svensson, I.; Sjöstedt-deLuna, S.; Bondesson, L., Estimation of wood fibre length distributions from censored data through an EM algorithm, Scand. J. statist., 33, 503-522, (2006) · Zbl 1114.62142
[25] Svensson, I.; Sjöstedt-deLuna, S.; Mörling, T.; Fries, A.; Ericsson, T., Adjusting for fibre length biased sampling probability using increment cores from standing trees, Holzforschung, 61, 101-103, (2007)
[26] Teicher, H., Identifiability of finite mixtures, Ann. math. statist., 34, 1265-1269, (1963) · Zbl 0137.12704
[27] van der Laan, M.J., Efficiency of the NPMLE in the line-segment problem, Scand. J. statist., 23, 527-550, (1996) · Zbl 0906.62032
[28] van der Laan, M.J., The two-interval line-segment problem, Scand. J. statist., 25, 163-186, (1998) · Zbl 0921.62038
[29] van Zwet, E.W., Laslett’s line segment problem, Bernoulli, 10, 377-396, (2004) · Zbl 1053.62047
[30] Wijers, B.J., Consistent non-parametric estimation for a one-dimensional line segment process observed in an interval, Scand. J. statist., 22, 335-360, (1995) · Zbl 0844.62027
[31] Wijers, B.J., 1997. Nonparametric estimation for a windowed line-segment process. Mathematical Center Tracts, vol. 121. Stichting Mathematisch Centrum, Amsterdam. · Zbl 0876.62029
[32] Wei, G.C.G.; Tanner, M.A., A Monte Carlo implementation of the EM algorithm and the poor Man’s data augmentation algorithms, J. amer. statist. assoc., 85, 699-704, (1990)
[33] Wu, C.F.J., On the convergence properties of the EM algorithm, Ann. statist., 11, 95-103, (1983) · Zbl 0517.62035
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. It attempts to reflect the references listed in the original paper as accurately as possible without claiming the completeness or perfect precision of the matching.