The motivation for this work is the desire to
e ciently solve systems of
nonlinear equations that have both linear and
nonlinear variables. This
type of system is sometimes referred to as a
separable system of
nonlinear equations. In this work, the Shen-Ypma
algorithm is
extended to overdetermined separable systems,
establishing the
connection with the Golub-Pereyra variable projection
method. A new
FORTRAN implementation of the method is provided and
early
developments in the field are surveyed.
e ciently solve systems of
nonlinear equations that have both linear and
nonlinear variables. This
type of system is sometimes referred to as a
separable system of
nonlinear equations. In this work, the Shen-Ypma
algorithm is
extended to overdetermined separable systems,
establishing the
connection with the Golub-Pereyra variable projection
method. A new
FORTRAN implementation of the method is provided and
early
developments in the field are surveyed.