USING PRINCIPAL-COMPONENTS REGRESSION TO STABILIZE EMG-MUSCLE FORCE PARAMETER ESTIMATES OF TORSO MUSCLES

Models for estimating muscle force from surface electromyographic (EMG ) recordings require parameter estimates with low intertrial variabili ty. The inclusion of multiple muscles in multivariate statistical mode ls can lead to multicollinearity, especially when there are significan t correlations between synergist muscles. One result of multicollinear ity is that parameter estimates are very sensitive to changes in the i ndependent variables. This study compared the parameter variability of multiple regression and principal-components regression techniques wh en applied to a six muscle EMG analysis of the lumbar region of the to rso, Nine subjects participated. Twenty-three percent of the tradition al multiple-regression parameters had incorrect signs, but none of the principal-components regression parameters did. The principal-compone nts regression technique also produced parameter estimates having an o rder of magnitude smaller parameter variability. It was concluded that principal-components regression is an effective method of mitigating the effect of multicollinearity in torso EMG models.