Abstract
Many stochastic process models for environmental data sets assume a process of relatively simple structure which is in some sense partially observed. That is, there is an underlying process (Xn, n ≥ 0) or (Xt, t ≥ 0) for which the parameters are of interest and physically meaningful, and an observable process (Yn, n ≥ 0) or (Yt, t ≥ 0) which depends on the X process but not otherwise on those parameters. Examples are wide ranging: the Y process may be the X process with missing observations; the Y process may be the X process observed with a noise component; the X process might constitute a random environment for the Y process, as with hidden Markov models; the Y process might be a lower dimensional function or reduction of the X process. In principle, maximum likelihood estimation for the X process parameters can be carried out by some form of the EM algorithm applied to the Y process data. In the paper we review some current methods for exact and approximate maximum likelihood estimation. We illustrate some of the issues by considering how to estimate the parameters of a stochastic Nash cascade model for runoff. In the case of k reservoirs, the outputs of these reservoirs form a k dimensional vector Markov process, of which only the kth coordinate process is observed, usually at a discrete sample of time points.
Similar content being viewed by others
References
Andrews, D.F.; Herzberg, A.M. 1985: Data. Springer-Verlag, New York
Baum, L.E. 1972: An inequality and associated maximization technique in statistical estimation for probabilistic functions of Markov processes, Inequalities 3, 1–8
Bickel, P.J.; Ritov, Y. 1993: Inference in hidden Markov models I: local asymptotic normality in the stationary case, Technical Report No. 383, Department of Statistics, University of California, Berkeley
Bodo, B.A.; Unny, T.E. 1987: On the outputs of the stochasticized Nash-Dooge linear reservoir cascade. Stochastic Hydrology IV, 131–147
Dempster, A.P.; Laird, N.M.; Rubin, D.B. 1977: Maximum likelihood from incomplete data via the EM algorithm, J. Royal Statist. Soc. Series B 39, 1–38
Fisher, R.A. 1925: Theory of statistical estimation. Proc. Camb. Phil. Soc. 22, 700–725
Gelfand, A.E.; Smith, A.F.M. 1990: Sampling-based approaches to calculating marginal densities, J. Amer. Statist. Assoc. 85, 398–409
Godambe, V.P. 1960: An optimal property of regular maximum likelihood estimation. Ann. Math. Statist. 31, 1208–1212
Godambe, V.P. 1985: The foundations of finite sample estimation in stochastic processes, Biometrika 72, 419–428
Kitigawa, G. 1987: Non-Gaussian state-space modeling of nonstationary time series. J. Amer. Statist. Assoc. 82, 1032–1041
Krishnamurthy, V.; Elliott, R.J. 1993: A parallel filtered-based EM algorithm for hidden Markov model and sinusoidal drift parameter estimation, 32nd IEEE Conference on Decision and Control. San Antonio, Texas, Dec. 15–17, 1993. IEEE Press, New York, 726–731
Küchler, U.; Sorensen, M. 1989: Exponential families of stochastic processes: A unifying semimartingale approach, International Statistical Review 57, 123–144
Leroux, B.C. 1992: Maximum likelihood methods for hidden Markov models. Stoch. Proc. and Appl. 40, 127–143
McLeish, D.L. 1983: Martingales and estimating equations for censored and aggregate data, Technical Report No. 12, Laboratory for Research in Statistics and Probability, Carleton University
McLeish, D.L. 1984: Estimation for aggregate models: The aggregate Markov chain, Canad. J. of Statist. 12, 265–282
Naik-Nimbalkar, U.V.; Rajarshi, M.B. 1992: Filtering and smoothing via estimating functions, In Recent Concepts in Statistical Inference, ed. J. Chen, Department of Statistics and Actuarial Science, University of Waterloo
Unny, T.E. 1987: Solutions to nonlinear stochastic differential equations in catchment modeling, Stochastic Hydrology IV, 87–111
Unny, T.E.; Karmeshu 1984: Stochastic nature of outputs from conceptual reservoir model cascades, J. Hydrol. 68, 161–180
West, M.P.; Harrison, P.J.; Migon, H.S. 1985: Dynamic generalized linear model and Bayesian forecasting, J. Amer. Statist. Assoc. 80, 732–796
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Thompson, M.E., Kaseke, T.N. Estimation for partially observed Markov processes. Stochastic Hydrol Hydraul 9, 33–47 (1995). https://doi.org/10.1007/BF01581757
Issue Date:
DOI: https://doi.org/10.1007/BF01581757