Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Nonlinear Time&hyphenSeries Prediction with Missing and Noisy Data

Tresp1, Hofmann

  • 1Siemens AG, Department of Information and Communications, Munich, DE, Otto-Hahn Ring 6, 81730. tresp@zfe.siemens.de

Neural Computation
|April 4, 1998
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

[Pregnancy and tumors of the pelvis minor].

Medisch maandblad·2014
Same author

Quantitative Model Studies on the Effectiveness of Different Precursor Systems in the Formation of the Intense Food Odorants 2-Furfurylthiol and 2-Methyl-3-furanthiol.

Journal of agricultural and food chemistry·2001
Same author

New and Convenient Syntheses of the Important Roasty, Popcorn-like Smelling Food Aroma Compounds 2-Acetyl-1-pyrroline and 2-Acetyltetrahydropyridine from Their Corresponding Cyclic alpha-Amino Acids.

Journal of agricultural and food chemistry·2001
Same author

Steady-state fully noninductive current driven by electron cyclotron waves in a magnetically confined plasma

Physical review letters·2000
Same author

Orientational ordering of N2O molecules adsorbed on graphite (0001): A novel commensurate pinwheel structure

Physical review letters·2000
Same author

"NaB15": a new structural description based on X-ray and neutron diffraction, electron microscopy, and solid-state NMR spectroscopy

Chemistry (Weinheim an der Bergstrasse, Germany)·2000
Same journal

A Model-Free Reinforcement Learning Implementation of Decision Making Under Uncertainty by Sequential Sampling.

Neural computation·2026
Same journal

DROP: Distributional and Regular Optimism and Pessimism for Reinforcement Learning.

Neural computation·2026
Same journal

Hierarchical Active Inference Using Successor Representations.

Neural computation·2026
Same journal

W-Kernel and Its Principal Space for Frequentist Evaluation of Bayesian Estimators.

Neural computation·2026
Same journal

A Hidden Markov Model-Inspired Sequence Classification Method for Hyperdimensional Computing.

Neural computation·2026
Same journal

Sparse Graphical Modeling for Electrophysiological Phase-Based Connectivity Using Circular Statistics.

Neural computation·2026
See all related articles

This study presents a probabilistic approach to handle missing and noisy data in nonlinear time series prediction. Stochastic simulation proves superior to simple iteration for K-step predictions, offering improved accuracy and error bar derivation.

Area of Science:

  • Nonlinear dynamics
  • Time series analysis
  • Probabilistic modeling

Background:

  • Missing and noisy data present significant challenges in accurate time series prediction.
  • Existing heuristic methods for data imputation can lead to suboptimal prediction outcomes.
  • Probabilistic frameworks offer a robust approach to address data uncertainties.

Purpose of the Study:

  • To develop and evaluate probabilistic solutions for nonlinear time series prediction with missing or noisy data.
  • To compare the efficacy of stochastic simulation against single-estimate imputation and predictor iteration.
  • To demonstrate the derivation of prediction error bars and applicability to K-step prediction.

Main Methods:

  • Derivation of prediction solutions from a probabilistic perspective.

Related Experiment Videos

  • Implementation and comparison of stochastic simulation and single-estimate imputation techniques.
  • Experimental validation using chaotic time series and real-world sunspot data.

    • Main Results:

    • Commonly used heuristics for data imputation were shown to yield suboptimal predictions.
    • Stochastic simulation demonstrated superior performance compared to simple predictor iteration for K-step prediction.
    • A method for deriving error bars for predictions was successfully demonstrated.

    Conclusions:

    • A probabilistic approach provides a robust framework for nonlinear time series prediction with data imperfections.
    • Stochastic simulation is a more effective method than predictor iteration for K-step forecasting.
    • The developed methods offer improved accuracy and uncertainty quantification in time series analysis.