Central Limit Theorems for Transition Probabilities of Controlled Markov Chains
By: Ziwei Su, Imon Banerjee, Diego Klabjan
Potential Business Impact:
Makes computers learn better from past decisions.
We develop a central limit theorem (CLT) for the non-parametric estimator of the transition matrices in controlled Markov chains (CMCs) with finite state-action spaces. Our results establish precise conditions on the logging policy under which the estimator is asymptotically normal, and reveal settings in which no CLT can exist. We then build upon it to derive CLTs for the value, Q-, and advantage functions of any stationary stochastic policy, including the optimal policy recovered from the estimated model. Goodness-of-fit tests are derived as a corollary, which enable us to test whether the logged data is stochastic. These results provide new statistical tools for offline policy evaluation and optimal policy recovery, and enable hypothesis tests for transition probabilities.
Similar Papers
CLT for LES of real valued random centrosymmetric matrices
Probability
Finds math patterns in complex data.
Statistical properties of Markov shifts (part I)
Probability
Finds patterns in changing random data.
Statistical properties of Markov shifts (part I)
Probability
Makes predictions about random events more accurate.