The problem of multidimensional nonlinear least-squares fitting requires the minimization of the squared residuals of n functions, f_i, in p parameters, x_i,
\Phi(x) = (1/2) || F(x) ||^2
= (1/2) \sum_{i=1}^{n} f_i(x_1, ..., x_p)^2
All algorithms proceed from an initial guess using the linearization,
\psi(p) = || F(x+p) || ~=~ || F(x) + J p ||
where x is the initial point, p is the proposed step and J is the Jacobian matrix J_{ij} = d f_i / d x_j. Additional strategies are used to enlarge the region of convergence. These include requiring a decrease in the norm ||F|| on each step or using a trust region to avoid steps which fall outside the linear regime.
To perform a weighted least-squares fit of a nonlinear model Y(x,t) to data (t_i, y_i) with independent gaussian errors \sigma_i, use function components of the following form,
f_i = (Y(x, t_i) - y_i) / \sigma_i
Note that the model parameters are denoted by x in this chapter since the non-linear least-squares algorithms are described geometrically (i.e. finding the minimum of a surface). The independent variable of any data to be fitted is denoted by t.
With the definition above the Jacobian is J_{ij} =(1 / \sigma_i) d Y_i / d x_j, where Y_i = Y(x,t_i).