Information Geometry Approach to Parameter Estimation in Markov Chains
arXiv:1401.3814 · doi:10.1214/15-AOS1420
Abstract
We consider the parameter estimation of Markov chain when the unknown transition matrix belongs to an exponential family of transition matrices. Then, we show that the sample mean of the generator of the exponential family is an asymptotically efficient estimator. Further, we also define a curved exponential family of transition matrices. Using a transition matrix version of the Pythagorean theorem, we give an asymptotically efficient estimator for a curved exponential family.
arXiv commentAppendix D is added
References in corpus (8)
- Curvature, concentration and error estimates for Markov chain Monte Carlo
- Explicit error bounds for lazy reversible Markov Chain Monte Carlo
- Nonasymptotic bounds on the estimation error of MCMC algorithms
- Rigorous confidence bounds for MCMC under a geometric drift condition
- Maximum Likelihood Estimation for Markov Chains
- Error bounds for computing the expectation by Markov chain Monte Carlo
- The exponential family of Markov chains and its information geometry
- Parallel Distributed Block Coordinate Descent Methods based on Pairwise Comparison Oracle
Cited by in corpus (10)
- Finite-length Analysis on Tail probability for Markov Chain and Application to Simple Hypothesis Testing
- Bregman divergence based em algorithm and its application to classical and quantum rate distortion theory
- Information Geometry of Reversible Markov Chains
- Universal channel coding for general output alphabet
- Information Geometry Approach to Parameter Estimation in Hidden Markov Models
- Local Equivalence Problem in Hidden Markov Model
- Asymptotic and Non-Asymptotic Analysis for Hidden Markovian Process with Quantum Hidden System
- The Cramér-Rao approach and global quantum estimation of bosonic states
- Legendre transformation and information geometry for the maximum entropy theory of ecology
- The space of positive transition measures on a Markov chain