Parameter Estimation in Matlab: Difference between revisions
From OpenWetWare
Jump to navigationJump to search
No edit summary |
(No difference)
|
Revision as of 07:36, 28 June 2005
Contrained Minimization
fmincon --> min f(x) over x for l<=x<=u
In this case we would define f(x) = MSE(sim(x, t), data(t)), where MSE is the mean squared error, and x is the set of adjustable parameters.
Nonlinear Curve Fitting
lsqcurvefit --> min 1/2|F(x,t)-y_data(t)|^2 over x for l<=x<=u