SIMULATING THE EFFECT OF MEASUREMENT ERROR ON STRIDE TIME DETRENDED FLUCTUATION ANALYSIS

Authors

Ben D. M. Jones, Sheffield Hallam University, La Trobe UniversityFollow
David L. Carey, La Trobe UniversityFollow
Jon Wheat, Nottingham Trent UniversityFollow
Kane Middleton, La Trobe UniversityFollow
Ben Heller, Sheffield Hallam UniversityFollow

Category

Modelling / Simulation

Document Type

Paper

Abstract

The aim of this study was to investigate the sensitivity of running stride time detrended fluctuation analysis (DFA) to simulated noise and the impact of noise on reliability measures. Varying magnitudes of Gaussian and uniform noise were added to series of touchdown event timings from 11 runners. DFA-α were calculated for ground truth and simulated stride time series. Simulated series were paired and test-retest ICC estimates computed. The error in DFA-α increased with noise magnitude, with stride time series transitioning from persistent to anti-persistent behaviour for greater magnitudes of noise. However, median ICC estimates stayed above 0.95 for all standard deviations of noise investigated. Thus, measurement noise in touchdown event timings is unlikely an explanation for the reported differences in the reliability of stride time DFA-α.

Recommended Citation

Jones, Ben D. M.; Carey, David L.; Wheat, Jon; Middleton, Kane; and Heller, Ben (2025) "SIMULATING THE EFFECT OF MEASUREMENT ERROR ON STRIDE TIME DETRENDED FLUCTUATION ANALYSIS," ISBS Proceedings Archive: Vol. 43: Iss. 1, Article 43.
Available at: https://commons.nmu.edu/isbs/vol43/iss1/43