This paper studies a parametric characterization of a dynamical human body. Inspired by systems medicine (SM), we propose to model the holistic human body as a linear time-invariant (LTI) system. The external environmental influences are viewed as input signals and multiple clinical measurements are regarded as output signals. In this work, six-channel wrist pulse signals measured by two pulse sensor arrays are taken as output signals to develop a vector autoregressive moving average with exogenous terms (VARMAX) model for characterizing an LTI human system. Once the model parameters are calculated from the measured signals, the LTI system is personalized. The personalized system can serve as a human body simulator, enabling one to predict clinical measurements such as the wrist pulse signals of a certain person. Experimental results show that the predicted wrist pulse signals are close enough to the actual measurements and using multiple complementary measurements can bring an improved performance.