Conditional exponential families and a representation theorem for asymptotic inference

DOI10.1214/aos/1176345463zbMath0476.62070OpenAlexW2058371490MaRDI QIDQ1159939

Paul D. Feigin

Publication date: 1981

Published in: The Annals of Statistics (Search for Journal in Brave)

Full work available at URL: https://doi.org/10.1214/aos/1176345463

zbMATH Keywords

asymptotic normalityadditive processesmaximum likelihood estimatemartingale propertyadditive exponential familyderivative of log likelihoodnonergodic stochastic processestime- homogeneous Markov chain

Mathematics Subject Classification ID

Markov processes: estimation; hidden Markov models (62M05) Markov processes (60J99)

Related Items

Large sample inference for conditional exponential families with applications to nonlinear time series, Exponential families of stochastic processes and Lévy processes, Inference and martingale estimating equations for stochastic processes on a semigroup, Asymptotic theory of conditional inference for stochastic processes, Information geometry approach to parameter estimation in Markov chains, Signal recovery by stochastic optimization, On exponential families of Markov processes, Longitudinal network models and permutation‐uniform Markov chains, Recursive parameter estimation: asymptotic expansion, On nonergodicity for nonparametric autoregressive models, Asymptotic optimal inference for a class of nonlinear time series models, Information geometry of reversible Markov chains, Stable convergence of semimartingales, Likelihood