ABSTRACT
Regression is a procedure for adjusting coefficient values in a mathematical model to have the model best fit the data. This chapter is an overview of the issues and a redirection to sources of techniques for nonlinear regression. There are many variants, but the common feature is that the objective function associated with classifications or ranks is discontinuous, not differentiable. The objective in nonlinear regression continues to be that of maximizing measures of goodness-of-fit by adjusting model coefficient values. Iterative techniques require a criterion for defining convergence – to identify when the optimizer is close enough to a solution, to determine when it can stop. Nonlinear regression means that the coefficients in the model do not appear linearly in the regression context, and/or that other nonlinearities cause difficulty in adjusting model coefficients.
