User`s guide

fitoptions
4-109
If Method is NonlinearLeastSquares, the additional fit option properties
shown below are available.
Property Description
Robust
Specifies whether to use the robust nonlinear least
squares fitting method. The value can be
{'off'} or 'on’.
Lower A vector of lower bounds on the coefficients to be fitted.
The default value is an empty vector indicating that the fit
is not constrained by lower bounds. If bounds are specified,
the vector length must equal the number of coefficients.
An unconstrained lower bound is specified by
-Inf.
Upper A vector of upper bounds on the coefficients to be fitted.
The default value is an empty vector indicating that the fit
is not constrained by upper bounds. If bounds are
specified, the vector length must equal the number of
coefficients. An unconstrained upper bound is specified by
Inf.
StartPoint Vector of coefficient starting values. The default value is
an empty vector. If the default value is passed to the
fit
function, then starting points for some library models are
determined heuristically. For other models, the values are
selected randomly on the interval (0,1).
Algorithm Algorithm used for the fitting procedure. The value can be
'Levenberg-Marquardt','Gauss-Newton', or
{'Trust-Region'}.
DiffMax
Change
Maximum change in coefficients for finite difference
gradients. The default value is 0.1.
DiffMin
Change
Minimum change in coefficients for finite difference
gradients. The default value is 10
-8
.
Display Level of display. {'notify'} displays output only if the fit
does not converge.
'final' displays only the final output.
'iter' displays output at each iteration. 'off' displays no
output.