Unbiased time-average estimators for Markov chains

2022/09/20 by Nabil Kahalé, Kahale, Nabil
Mathematics · Computer Science · #Statistical Methods and Inference #Bayesian Methods and Mixture Models #Statistical Methods and Bayesian Inference

paper · pdf · doi:10.48550/arxiv.2209.09581

Abstract

We consider a time-average estimator f_k of a functional of a Markov chain. Under a coupling assumption, we show that the expectation of f_k has a limit μ as the number of time-steps goes to infinity. We describe a modification of f_k that yields an unbiased estimator f_k of μ. It is shown that f_k is square-integrable and has finite expected running time. Under certain conditions, f_k can be built without any precomputations, and is asymptotically at least as efficient as f_k, up to a multiplicative constant arbitrarily close to 1. Our approach provides an unbiased estimator for the bias of f_k. We study applications to volatility forecasting, queues, and the simulation of high-dimensional Gaussian vectors. Our numerical experiments are consistent with our theoretical findings.

Related

• Exact Estimation for Markov Chain Equilibrium Expectations
• Solving the Poisson equation using coupled Markov chains
• Optimistic Estimation of Convergence in Markov Chains with the Average-Mixing Time
• Approximating Performance Measures for Slowly Changing Non-stationary Markov Chains
• Robust estimation of a Markov chain transition matrix from multiple sample paths