Limits to classification and regression estimation from ergodic processes

DOI10.1214/aos/1018031110zbMath0933.62033OpenAlexW1571830002MaRDI QIDQ1807170

Andrew B. Nobel

Publication date: 9 November 1999

Published in: The Annals of Statistics (Search for Journal in Brave)

Full work available at URL: https://doi.org/10.1214/aos/1018031110

zbMATH Keywords

counterexamplesergodic processesreduction arguments

Mathematics Subject Classification ID

Nonparametric regression and quantile regression (62G08) Density estimation (62G07) Stationary stochastic processes (60G10) Inference from stochastic processes (62M99)

Related Items

Hypothesis testing for families of ergodic processes ⋮ ON SEQUENTIAL ESTIMATION AND PREDICTION FOR DISCRETE TIME SERIES ⋮ Fast learning from \(\alpha\)-mixing observations ⋮ Learning Theory Estimates with Observations from General Stationary Stochastic Processes ⋮ On universal estimates for binary renewal processes ⋮ Learning from dependent observations ⋮ On universal algorithms for classifying and predicting stationary processes ⋮ Consistency of support vector machines for forecasting the evolution of an unknown ergodic dynamical system from observations with unknown noise ⋮ Unnamed Item

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Prediction of random sequences and universal coding
• Distribution-free consistency results in nonparametric discrimination and regression function estimation
• Consistent window estimation in nonparametric regression
• Strong consistent density estimate from ergodic sample
• Nonparametric curve estimation from time series
• Universal schemes for prediction, gambling and portfolio selection
• Consistent nonparametric regression. Discussion
• Guessing the next output of a stationary process
• On density estimation from ergodic processes
• Nearest neighbor regression estimation for null-recurrent Markov time series
• On the strong uniform consistency of density estimation for strongly dependent sequences
• Strongly consistent nonparametric forecasting and regression for stationary ergodic sequences.
• Correction to: Universal prediction schemes
• Nonparametric inference for ergodic, stationary time series
• Nonparametric estimation in Markov processes
• Kernel density estimation from ergodic sample is not universally consistent
• NON‐PARAMETRIC ESTIMATION WITH STRONGLY DEPENDENT MULTIVARIATE TIME SERIES
• MULTIVARIATE LOCAL POLYNOMIAL REGRESSION FOR TIME SERIES:UNIFORM STRONG CONSISTENCY AND RATES
• Weakly convergent nonparametric forecasting of stationary time series
• Nonparametric estimation of a regression function and its derivatives under an ergodic hypothesis
• Density estimation from an individual numerical sequence
• Limits to consistent on-line forecasting for ergodic time series
• Regression Estimation from an Individual Stable Sequence
• Nonparametric Estimation of the Transition Distribution Function of a Markov Process