Subset Selection for Improved Parameter Estimation

In this paper we discuss methods for subset selection for a non-linear least squares parameter estimation problem. Through the use of these algorithms, we partition the parameter space into a linear independent and a linearly dependent set. By identifying these subsets, we are able to gain further into model dynamics. We then apply subset selection techniques to a simplified model describing the physiologic response of the human immunodeficiency virus (HIV) to identify the most-identifiable parameters.