Handy sufficient conditions for the convergence of the maximum likelihood estimator in observation-driven models

DOI10.1007/s10986-015-9286-8zbMath1323.62027arXiv1506.01831OpenAlexW1538855836MaRDI QIDQ746977

Randal Douc, Tepmony Sim, François Roueff

Publication date: 22 October 2015

Published in: Lithuanian Mathematical Journal (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/1506.01831

zbMATH Keywords

consistencyergodicitymaximum likelihoodtime series of countsobservation-driven models

Mathematics Subject Classification ID

Asymptotic properties of parametric estimators (62F12) Discrete-time Markov processes on general state spaces (60J05)

Related Items

Estimation of agent-based models using sequential Monte Carlo methods, General-order observation-driven models: ergodicity and consistency of the maximum likelihood estimator, COUNT AND DURATION TIME SERIES WITH EQUAL CONDITIONAL STOCHASTIC AND MEAN ORDERS, Necessary and sufficient conditions for the identifiability of observation‐driven models

Cites Work

• Unnamed Item
• The maximizing set of the asymptotic normalized log-likelihood for partially observed Markov chains
• On weak dependence conditions for Poisson autoregressions
• Log-linear Poisson autoregression
• Absolute regularity and ergodicity of Poisson count processes
• On recursive estimation for time varying autoregressive processes
• Maximum-likelihood estimation for hidden Markov models
• Generalized autoregressive conditional heteroscedasticity
• Stationarity of generalized autoregressive moving average models
• Ergodicity of observation-driven time series models and consistency of the maximum likelihood estimator
• On the measurability and consistency of minimum contrast estimates
• Observation-driven models for Poisson counts
• Poisson Autoregression
• Markov Chains and Stochastic Stability
• A negative binomial integer-valued GARCH model