Constant term in fittype

Hello, i want to fit some data with the funktion: y= S0/2+S1/2*cos(2*x)+S2/2*sin(2*x) while i know the value of S0 and the values of S1 and S2 should be fitted by matlab.

So my current matlab code is

        poincarefit=fittype({'S0/2+S1/2*cos(2*x)+S2/2*sin(2*x)'},...
                             'problem','S0');
        f = fit(Theta, I, poincarefit,'problem',S0);

But that throws me the error

    Expression a*(S0/2+S1/2*cos(2*x)+S2/2*sin(2*x)) is not a valid MATLAB expression, has non-scalar
    coefficients, or cannot be evaluated:
    Error in fittype linear term ==> (S0./2+S1./2.*cos(2.*x)+S2./2.*sin(2.*x))
    ??? Undefined function or variable 'S1'.
end

so i tried to implement the coefficients like this:

        poincarefit=fittype({'S0/2+S1/2*cos(2*x)+S2/2*sin(2*x)'},...
                    'coefficients',{'S1','S2'},'problem','S0');
        f = fit(Theta, I, poincarefit,'problem',S0);

but the error massage

Number of coefficients must match the number of linear terms.

occures...

can you maybe help me?

thanks in advance

Answer 1

[Ganesan S]

Dear Sir,

can you try this?

x=0:.1:2*pi;    % Whatever your x data or time
y= cos(2*x)+sin(2*x);    % Whatever your y data
[xData, yData] = prepareCurveData( x, y);
S0=1;           % you said you know this number
eqn = @(S1,S2,x) S0/2+S1/2*cos(2*x)+S2/2*sin(2*x); % a different format instead of typing the equation in fittype
opts = fitoptions( 'Method', 'NonlinearLeastSquares' );
opts.StartPoint = [1 1];%starting guess for S1 and S2
[fitresult, gof] = fit( xData, yData, eqn, opts );
% Plot fit with data.
figure( 'Name', 'untitled fit 1' );
h = plot( fitresult, xData, yData );
legend( h, 'y vs. x', 'untitled fit 1', 'Location', 'NorthEast' );
% Label axes
xlabel x
ylabel y
grid on

I hope this helps... I am really not sure it helps you as you have posted this question in the year 2015 but now it is end of 2018. I answered this because this answer is also needed for me! When I searched for an answer to this, it is looking very strange that nobody had answered your query. All the best!

Yours sincerely,

S.Ganesan.