Abstract
Information inequalities in a general sequential model for stochastic processes are presented by applying the approach to estimation through estimating functions. Using this approach, Bayesian versions of the information inequalities are also obtained. In particular, exponential-family processes and counting processes are considered. The results are useful to find optimum properties of parameter estimators. The assertions are of great importance for describing estimators in failure-repair models in both Bayes approach and the nuisance parameter case.
Similar content being viewed by others
References
Andersen PK, Borgan Ø, Gill RD, Keiding N (1993) Statistical models based on counting processes. Springer, New York-London
Bain LJ, Engelhardt M (1986) On the mean time between failures for repairable systems. IEEE transac. on Reliability R-35/4:419–422
Chang IS, Hsiung CA (1990) Time-sequential point estimation through estimating equations. Ann. Statist. 18:1378–1388
Cohen AC (1965) Maximum likelihood estimation in Weibull distribution based on censored samples. Technometrics 7/4:579–588
Cox DR, Oakes D (1984) Analysis of survival data. Chapman and Hall, London-New York
Döhler R (1981) Dominierberkeit und Suffizienz in der Sequentialanalyse. Math. Operations-forsch. Statist. Ser. Statist. 12:101–134
Ferreira PE (1982) Sequential estimation through estimating equations in the nuisance parameter case. Ann. Statist. 10:167–173
Franz J (1982) Sequential estimation and asymptotic properties in birth-and-death processes. Math. Operationsforsch. Statist., Ser. statistics 13/2:231–244
Franz J (1994) On estimation problems in random censored repair models. Economic Quality Control 9/3:125–142
Franz J, Magiera R (1990) Admissible estimators in sequential plans for exponential-type processes. Scand. J. Statist. 17:275–285
Gill RD, Levit BY (1995) Applications of the van Trees inequality: a Bayesian Cramér-Rao bound. Bernoulli 1 (1/2): 59–79
Godambe VP (1960) An optimum property of regular maximum likelihood estimation. Ann. Math. Statist. 31:1208–1212
Godambe VP (1976) Conditional likelihood and unconditional optimum estimating functions. Biometrika 63:277–284
Godambe VP, Thompson ME (1974) Estimating equations in the presence of a nuisance parameter. Ann. Statist. 2:568–574
Küchler U, Sørensen M (1994) Exponential families of stochastic processes and Lévy processes. J. Statist. Plan. Infer. 39:211–237
Liptser RS, Shiryaev AN (1978) Statistics of random processes, Vol. II. Springer, New York-London
Magiera R (1974) On the inequality of Cramér-Rao type in sequential estimation theory. Zastos. Mat. 14:227–235
Magiera R (1994a) Sequential estimation through estimating equations. In: Bachem et al. (eds) Operations Research '93, Proc. 18th Symp. Operat. Res, Physica Verlag, Heidelberg: 335–338
Magiera R (1994b) Conjugate priors for exponential-type processes with random initial conditions. Applicationes Mathematicae 22(3):321–330
Magiera R, Stefanov VT (1989) Sequential estimation in exponential-type processes under random initial conditions. Sequential Analysis 8/2:147–167
Mann NR, Schafer RE, Singpurwalla ND (1974) Methods for statistical analysis of reliability and life data. Wiley New York-London-Toronto
Pruscha H (1984) Statistical inference in multivariate point processes with parametric intensity. Preprint Univ. München, Math. Institute
Stefanov VT (1991) A note on the moments of first-crossing times related to nonlinear renewal theory. Stochastics and Stochastics Reports 35:25–30
Sudakov VN (1969) On measures defined by Markovian moments. Investigations on the Theory of Random Processes, Vol. 12, Memoirs of the Scientific Seminars of the Leningrad Section of the Steklov Math. Inst.: 157–164 (in Russian)
Wolfowitz J (1947) The efficiency of sequential estimates and Wald's equation for sequential processes. Ann. Math. Statist. 18:215–230
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Franz, J., Magiera, R. On information inequalities in sequential estimation for stochastic processes. Mathematical Methods of Operations Research 46, 1–27 (1997). https://doi.org/10.1007/BF01199461
Issue Date:
DOI: https://doi.org/10.1007/BF01199461