Turning Point Prediction of Oscillating Time Series using Local Dynamic Regression Models

TL;DR

This paper introduces a local dynamic regression approach focused on predicting the turning points of oscillating time series, improving accuracy over standard models by reconstructing state space using only turning points.

Contribution

It extends existing local models by applying them to turning points, enhancing prediction accuracy for oscillating nonlinear systems.

Findings

Better turning point prediction accuracy than standard local models

Effective on nonlinear oscillating systems

Improved forecasting of oscillation characteristics

Abstract

In the prediction of oscillating time series, the interest is in the turning points of successive oscillations rather than the samples themselves. For this purpose a scheme has been proposed; the state space reconstruction is limited to the turning points and the local (nearest neighbor) model is modified in order to predict the turning point magnitudes and times. This approach is extended here using a local dynamic regression model on both turning point magnitudes and times. Simulations on oscillating nonlinear systems show that the proposed approach gives better predictions of turning points than the standard local model applied to all the samples of the oscillating time series.