Utilizing an ARIMA Model to Assess Gait Stability and Spectral Behavior

• Activity Number: 184
• Type: Contributed
• Date/Time: Monday, August 10, 2015 : 10:30 AM to 12:20 PM
• Sponsor: Section on Statistics in Sports
• Author(s): Kristin Denise Morgan* and Brian Noehren
• Companies: University of Kentucky and University of Kentucky
• Keywords: ARIMA ; Time Series ; Stability ; Anterior Cruciate Ligament Injuries ; Data Mining

Abstract

Past research revealed a novel technique (Bode stability criteria) to evaluate an individual's stability while performing walking and/or running protocols. The technique employed an output transfer function that was a response to a perturbed stimulus induced during the initial phase of the captured dynamic motion. In this study, sagittal plane kinematic knee waveform data were captured on thirty-two control and post Anterior Cruciate Ligament (ACL) reconstruction participants. The coefficients from an autoregressive integrated moving average (ARIMA) model fit to the data from the walking and running protocols were employed to assess individual gait stability, derive directional stability vectors and estimate the degree of stability separation among the individuals and population groups. The findings from this investigation will replace the cumbersome and time-consuming previously employed curve fitting technique with a stability transfer function that can be used to assess the spectral nature of the derived ARIMA models.