Extracting values from a smoothing spline fit

I've created a spline fit using curve fitting

xdat = [1;2;3;4;5;6;7;8;9;10]
ydat = [2;3;5;7;9;14;24;46;57;58]

SplineFit = fit(xdat, ydat, 'smoothingspline');

I can plot this using simply

plot(fitA)

However, what I really want to do is use this plot to find the y values for x values that are not input x values. Sort of an interpolation, but using the spline fit I used.

i.e. I want to find ydat values where xdat=[1.2; 3.6; 6.2; 7.8], using SplineFit

If I were using polynomials, I know I would use the polyval function for this but I don't know what to use considering I've used a smoothing spline.

Any help appreciated

Answer 1

[John D'Errico]

feval(SplineFit,xdat)
ans =
       2.1463
       5.9731
       15.993
       40.672

You could have known this by looking at what was returned. What is it?

whos SplineFit
  Name           Size            Bytes  Class    Attributes

    SplineFit      1x1              2088  cfit               

>> help cfit
  cfit   Curve Fit Object

      A cfit object encapsulates the result of fitting a curve to data.
      Use the FIT function or CFTOOL to create cfit objects.

      You can treat a cfit object as a function to make predictions or
      evaluation the curve at values of X. For example to predict what
      the population was in 1905, then use

          load census
          cf = fit( cdate, pop, 'poly2' );
          yhat = cf( 1905 )

      cfit methods:
          coeffvalues   - Get values of coefficients.
          confint       - Compute prediction intervals for coefficients.
          differentiate - Compute values of derivatives.
          feval         - Evaluate at new observations.
          integrate     - Compute values of integral.
          plot          - Plot a curve fit.
          predint       - Compute prediction intervals for a curve fit or for 
                          new observations.
          probvalues    - Get values of problem parameters.

      See also: fit, fittype, sfit.

      Reference page in Help browser
         doc cfit

>> methods cfit

Methods for class cfit:

argnames       coeffnames     dependnames    fitoptions     integrate      numcoeffs      probnames      type           
category       coeffvalues    differentiate  formula        islinear       plot           probvalues     
cfit           confint        feval          indepnames     numargs        predint        setoptions

In that list of methods, you would have seen the feval method, clearly what you need.

You would also have seen that SplineFit itself can be used as a function. Thus...

SplineFit(xdat)
ans =
       2.1463
       5.9731
       15.993
       40.672

So, read the help! It tells you a lot.

[Mat]

Thanks John, I had browsed through help docs online but had obviously focused my attentions in the wrong places (e.g. help fit). Didn't know/think to search for cfit - the output variable - I had been searching for around an hour for this function that someone else might know off-hand, so it wasn't for lack of trying! Thanks again.

[John D'Errico]

True. Things are not always obvious.

[Mohamed]

May i please know how to use the cfit object in the simulink interface to predict output for random input values ?

Answer 2

[Image Analyst]

I think that's what my attached demo does.