Computational Limits to Nonparametric Estimation for Ergodic Processes
From MaRDI portal
Abstract: A new negative result for nonparametric estimation of binary ergodic processes is shown. I The problem of estimation of distribution with any degree of accuracy is studied. Then it is shown that for any countable class of estimators there is a zero-entropy binary ergodic process that is inconsistent with the class of estimators. Our result is different from other negative results for universal forecasting scheme of ergodic processes.
Recommendations
- Nonparametric Statistical Inference for Ergodic Processes
- Limits to classification and regression estimation from ergodic processes
- On stochastic complexity and nonparametric density estimation
- Nonparametric prediction from ergodic samples
- Nonparametric inference for ergodic, stationary time series
- scientific article; zbMATH DE number 4052721
- scientific article; zbMATH DE number 66847
Cited in
(6)- On universal algorithms for classifying and predicting stationary processes
- Universal rates for estimating the residual waiting time in an intermittent way.
- Consistency, integrability and asymptotic normality for some intermittent estimators
- Inferring the residual waiting time for binary stationary time series
- scientific article; zbMATH DE number 7639721 (Why is no real title available?)
- Countable alphabet stationary processes with at least one memory word and intermittent estimation with universal rates
This page was built for publication: Computational Limits to Nonparametric Estimation for Ergodic Processes
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5272319)
