Learning and Long-run Fundamentals in Stationary Environments

Research output: Working paper

Abstract

We prove that a Bayesian agent observing a stationary process learns to make predictions as if the fundamental, defined as the ergodic component or the long run empirical frequencies of the process, was known to him. We interpret the ergodic representation as a decomposition of a stationary belief into risk and uncertainty.
Original languageEnglish (US)
Number of pages22
StatePublished - Dec 23 2013

Fingerprint Dive into the research topics of 'Learning and Long-run Fundamentals in Stationary Environments'. Together they form a unique fingerprint.

Cite this