Recursive construction of confidence regions

DOI10.1214/17-EJS1362MaRDI QIDQ1684156zbMATH OpenOpenAlexFDO

Authors Tao Chen, Igor Cialenco, Tomasz R. Bielecki

Publication date 8 December 2017

Published in Electronic Journal of Statistics (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1605.08010, https://projecteuclid.org/euclid.ejs/1510974130

zbMATH Keywords

ergodic processesquasi-asymptotically linear estimatorrecursive confidence regionsrecursive point estimatorsstatistical inference for Markov chainsstochastic approximation

Mathematics Subject Classification ID

Discrete-time Markov processes on general state spaces (60J05) Applications of Markov chains and discrete-time Markov processes on general state spaces (social mobility, learning theory, industrial processes, etc.) (60J20) Point estimation (62F10) Asymptotic properties of parametric estimators (62F12) Parametric tolerance and confidence regions (62F25) Markov processes: estimation; hidden Markov models (62M05)

Abstract: Assuming that one-step transition kernel of a discrete time, time-homogenous Markov chain model is parameterized by a parameter $$, we derive a recursive (in time) construction of confidence regions for the unknown parameter of interest, say $$. It is supposed that the observed data used in construction of the confidence regions is generated by a Markov chain whose transition kernel corresponds to $het{a}^{\ast }$ . The key step in our construction is derivation of a recursive scheme for an appropriate point estimator of $het{a}^{\ast }$. To achieve this, we start by what we call the base recursive point estimator, using which we design a quasi-asymptotically linear recursive point estimator (a concept introduced in this paper). For the latter estimator we prove its weak consistency and asymptotic normality. The recursive construction of confidence regions is needed not only for the purpose of speeding up the computation of the successive confidence regions, but, primarily, for the ability to apply the dynamic programming principle in the context of robust adaptive stochastic control methodology.

Recommendations

• Accurate confidence regions for Markov-model parameters
• Recursive estimation: asymptotic confidence regions by empirical quantiles
• Sequential confidence regions for maximum likelihood estimates.
• Markov chain Monte Carlo confidence intervals
• scientific article; zbMATH DE number 3896107

Cited in

(12)

• Online drift estimation for jump-diffusion processes
• Gittins' theorem under uncertainty
• Accurate confidence regions for Markov-model parameters
• Recursive confidence band construction for an unknown distribution function
• Adaptive robust control under model uncertainty
• Time-inconsistent Markovian control problems under model uncertainty with application to the mean-variance portfolio selection
• Risk-sensitive Markov decision problems under model uncertainty: finite time horizon case
• Adaptive robust control in continuous time
• A machine learning approach to adaptive robust utility maximization and hedging
• Nonparametric Adaptive Robust Control under Model Uncertainty
• Portfolio diversification and model uncertainty: A robust dynamic mean‐variance approach
• Data-driven nonparametric robust control under dependence uncertainty

This page was built for publication: Recursive construction of confidence regions

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1684156)