User`s guide

fit
4-104
example, the information returned for nonlinear least squares fits is given
below.
Remarks For rationals and Weibull library models, the coefficient starting values are
randomly selected in the range [0,1]. Therefore, if you perform multiple fits to
a data set using the same equation, you might get different coefficient results
due to different starting values. To avoid this situation, you should pass in a
vector of starting values each time you fit, or define a specific state for the
random number generator,
rand or randn, before fitting.
For all other library models, optimal starting points are automatically
calculated. These values depend on the data, and are based on model-specific
heuristics.
Field Description
numobs
Number of observations (response values).
numparam Number of unknown parameters to fit.
residuals Vector of residuals.
Jacobian Jacobian matrix.
exitflag Describes the exit condition. If exitflag > 0, the
function converged to a solution. If
exitflag = 0, the
maximum number of function evaluations or iterations
was exceeded. If
exitflag < 0, the function did not
converge to a solution.
iterations Number of iterations used to complete the fit.
funcCount Number of function evaluations used to complete the fit.
firstorderopt Measure of first-order optimality.
algorithm Fitting algorithm used.