MODELING OF ADAPTATIONS TO PHYSICAL-TRAINING BY USING A RECURSIVE LEAST-SQUARES ALGORITHM

The present study assesses the usefulness of a systems model with time -varying parameters for describing the responses of physical performan ce to training. Data for two subjects who undertook a 14-wk training o n a cycle ergometer were used to test the proposed model, and the resu lts were compared with a model with time-invariant parameters. Two 4-w k periods of intensive training were separated by a 2-wk period of red uced training and followed by a 4-wk period of reduced training. The s ystems input ascribed to the training doses was made up of interval ex ercises and computed in arbitrary units. The systems output was evalua ted one to five times per week by using the endurance time at a consta nt workload. The time-invariant parameters were fitted from actual per formances by using the least squares method. The time-varying paramete rs were fitted by using a recursive least squares algorithm. The coeff icients of determination r(2) were 0.875 and 0.879 for the two subject s using the time-varying model, higher than the values of 0.682 and 0. 666, respectively, obtained with the time-invariant model. The variati ons over time in the model parameters resulting from the expected redu ction in the residuals appeared generally to account for changes in re sponses to training. Such a model would be useful for investigating th e underlying mechanisms of adaptation and fatigue.