Prediction of the hourly ionospheric parameter f(O)F(2) using a novel nonlinear interpolation technique to cope with missing data points

N M Francis, A G Brown, P S Cannon, D S Broomhead

Research output: Contribution to journalArticle

Abstract

A technique for neural network time series prediction using radial basis functions, where the input data contain a significant proportion of missing points, is developed. This technique is intended to model the data while simultaneously providing a means of minimizing the impact upon the model of the missing points that are typical of geophysical time series. The two issues are inextricably entwined because missing data points have a significant impact upon the performance of data-derived models in terms of prediction availability and accuracy. The core of the technique is a nonlinear interpolation scheme that assigns values to gaps in the input time series. Each missing point is interpolated such that the error introduced into any specific predictive function is minimized. This interpolative technique has a general application in any instance where the effects of interpolation upon a given analysis process need to be minimized or a complete time series needs to be constructed from incomplete data. The technique has been applied to the prediction of f(O)F(2) from Slough, United Kingdom. The resultant model prediction root-mean-square (RMS) error is shown to be 2.3% better than using recurrence interpolation (in terms of overall model accuracy rather than relative to each other), 3.8% better than using persistence interpolation, and 34.3% better than not using any interpolation. Utilizing the interpolation algorithm lowers the RMS error by 26% when incomplete data, in addition to complete data, are used as an input to both the interpolated and the uninterpolated models.
Original languageEnglish
Pages (from-to)30077-30083
Number of pages7
JournalJournal of Geophysical Research
Volume106
Issue numberA12
Publication statusPublished - 2001

Fingerprint

ionospherics
interpolation
Interpolation
prediction
Time series
time series analysis
predictions
time series
root-mean-square errors
Mean square error
methodology
United Kingdom
neural networks
availability
parameter
proportion
Availability
Neural networks
persistence

Cite this

Prediction of the hourly ionospheric parameter f(O)F(2) using a novel nonlinear interpolation technique to cope with missing data points. / Francis, N M; Brown, A G; Cannon, P S; Broomhead, D S.

In: Journal of Geophysical Research, Vol. 106, No. A12, 2001, p. 30077-30083.

Research output: Contribution to journalArticle

@article{dbd6ecaf688146269b531a7e046c2dd4,
title = "Prediction of the hourly ionospheric parameter f(O)F(2) using a novel nonlinear interpolation technique to cope with missing data points",
abstract = "A technique for neural network time series prediction using radial basis functions, where the input data contain a significant proportion of missing points, is developed. This technique is intended to model the data while simultaneously providing a means of minimizing the impact upon the model of the missing points that are typical of geophysical time series. The two issues are inextricably entwined because missing data points have a significant impact upon the performance of data-derived models in terms of prediction availability and accuracy. The core of the technique is a nonlinear interpolation scheme that assigns values to gaps in the input time series. Each missing point is interpolated such that the error introduced into any specific predictive function is minimized. This interpolative technique has a general application in any instance where the effects of interpolation upon a given analysis process need to be minimized or a complete time series needs to be constructed from incomplete data. The technique has been applied to the prediction of f(O)F(2) from Slough, United Kingdom. The resultant model prediction root-mean-square (RMS) error is shown to be 2.3{\%} better than using recurrence interpolation (in terms of overall model accuracy rather than relative to each other), 3.8{\%} better than using persistence interpolation, and 34.3{\%} better than not using any interpolation. Utilizing the interpolation algorithm lowers the RMS error by 26{\%} when incomplete data, in addition to complete data, are used as an input to both the interpolated and the uninterpolated models.",
author = "Francis, {N M} and Brown, {A G} and Cannon, {P S} and Broomhead, {D S}",
note = "ID number: ISI:000173728500105",
year = "2001",
language = "English",
volume = "106",
pages = "30077--30083",
journal = "Journal of Geophysical Research",
issn = "0148-0227",
publisher = "American Geophysical Union",
number = "A12",

}

TY - JOUR

T1 - Prediction of the hourly ionospheric parameter f(O)F(2) using a novel nonlinear interpolation technique to cope with missing data points

AU - Francis, N M

AU - Brown, A G

AU - Cannon, P S

AU - Broomhead, D S

N1 - ID number: ISI:000173728500105

PY - 2001

Y1 - 2001

N2 - A technique for neural network time series prediction using radial basis functions, where the input data contain a significant proportion of missing points, is developed. This technique is intended to model the data while simultaneously providing a means of minimizing the impact upon the model of the missing points that are typical of geophysical time series. The two issues are inextricably entwined because missing data points have a significant impact upon the performance of data-derived models in terms of prediction availability and accuracy. The core of the technique is a nonlinear interpolation scheme that assigns values to gaps in the input time series. Each missing point is interpolated such that the error introduced into any specific predictive function is minimized. This interpolative technique has a general application in any instance where the effects of interpolation upon a given analysis process need to be minimized or a complete time series needs to be constructed from incomplete data. The technique has been applied to the prediction of f(O)F(2) from Slough, United Kingdom. The resultant model prediction root-mean-square (RMS) error is shown to be 2.3% better than using recurrence interpolation (in terms of overall model accuracy rather than relative to each other), 3.8% better than using persistence interpolation, and 34.3% better than not using any interpolation. Utilizing the interpolation algorithm lowers the RMS error by 26% when incomplete data, in addition to complete data, are used as an input to both the interpolated and the uninterpolated models.

AB - A technique for neural network time series prediction using radial basis functions, where the input data contain a significant proportion of missing points, is developed. This technique is intended to model the data while simultaneously providing a means of minimizing the impact upon the model of the missing points that are typical of geophysical time series. The two issues are inextricably entwined because missing data points have a significant impact upon the performance of data-derived models in terms of prediction availability and accuracy. The core of the technique is a nonlinear interpolation scheme that assigns values to gaps in the input time series. Each missing point is interpolated such that the error introduced into any specific predictive function is minimized. This interpolative technique has a general application in any instance where the effects of interpolation upon a given analysis process need to be minimized or a complete time series needs to be constructed from incomplete data. The technique has been applied to the prediction of f(O)F(2) from Slough, United Kingdom. The resultant model prediction root-mean-square (RMS) error is shown to be 2.3% better than using recurrence interpolation (in terms of overall model accuracy rather than relative to each other), 3.8% better than using persistence interpolation, and 34.3% better than not using any interpolation. Utilizing the interpolation algorithm lowers the RMS error by 26% when incomplete data, in addition to complete data, are used as an input to both the interpolated and the uninterpolated models.

M3 - Article

VL - 106

SP - 30077

EP - 30083

JO - Journal of Geophysical Research

JF - Journal of Geophysical Research

SN - 0148-0227

IS - A12

ER -