MATLAB: Does FITTYPE returns ‘cfit’ when passed a cfit object when using the Curve Fitting Toolbox

cfitcurveCurve Fitting Toolboxfitfittypemodel

FITTYPE returns 'cfit' when passed a cfit object when using the Curve Fitting Toolbox.

Best Answer

This bug has been fixed in Release 14 Service Pack 2 (R14SP2). For previous releases, there are no workarounds.

Related Solutions

MATLAB: How to obtain the coefficient values, confidence intervals, and other data generated by the CFTOOL GUI within the Curve Fitting Toolbox from the saved fitted model object

This enhancement has been applied for Release 14 SP1 (R14SP1). For previous releases, read below for any possible workarounds:

To obtain the coefficient values from a "cfit" object, use the COEFFVALUES function. For example, if you saved a fitted model from the CFTOOL to a variable named "MyModel", the following command will extract the coefficient values:

MyCoeffs = coeffvalues(MyModel)

For example, the second coefficient "p2" (if it exists) can be obtained with:

coeff_2 = MyCoeffs(2)

You can also use several other functions within the Curve Fitting Toolbox to extract additional information from the "cfit" object.

 CONFINT produces the confidence intervals on the coefficients.
 DIFFERENTIATE differentiates the fit.
 INDEPNAMES extracts the names of the independent parameters of the fit.
 INTEGRATE integrates the fit.
 PLOT plots the fit.
 PREDINT computes confidence intervals on the predictions produced by the fit.

All these functions are used in a manner similar to COEFFVALUES. Refer to the documentation for the exact syntax for each of these functions.

MATLAB: How to specify a model asa FITTYPE object and not as a function using the Curve Fitting Toolbox

You can specify a FITTYPE object as your fit model in FIT function.

Use an object FITTYPE (different from CFIT). Here is an example which demsonstrates how to do this:

   f1 = fittype('a*x+b')
 f1 =
      General model:
        f1(a,b,x) = a*x+b

Now, you can use this model to fit different data set. For example:

x = 0:10; y = sin(x);
f2 = fit(x',y',f1)
 f2 =
   General model:
     f2(x) = a*x+b
   Coefficients (with 95% confidence bounds):
     a =    -0.02444  (-0.1812, 0.1323)
     b =      0.2505  (-0.677, 1.178)

Best Answer

Related Solutions

MATLAB: How to obtain the coefficient values, confidence intervals, and other data generated by the CFTOOL GUI within the Curve Fitting Toolbox from the saved fitted model object

MATLAB: How to specify a model asa FITTYPE object and not as a function using the Curve Fitting Toolbox

Related Question