MATLAB: Question about multi variable data fitting

Question

Hi,

I am trying to do a simple two variable data fitting, using fittype:

g=fittype('a*exp(-(x-b)^2/c^2-(y-d)^2/f^2)','independent', {'x' 'y'});

However, everytime I run the program, error message says:

Error using internal.matlab.Message In 'curvefit:fittype:SameNameIndAndDepenVars', parameter {1} must be a scalar.

Error in message (line 11) msgObj = internal.matlab.Message(msgID,varargin{:});

Error in fittype>iTestCustomModelParameters (line 811) error(message('curvefit:fittype:SameNameIndAndDepenVars', obj.indep));

Error in fittype (line 355) iTestCustomModelParameters( obj );

Error in LED_roll (line 13) g=fittype('a*exp(-(x-b)^2/c^2-(y-d)^2/f^2)','independent', {'x' 'y'});

Please let me know where I did wrong here.

Thanks

Answer 1

Give this a try:

g=fittype('a*exp(-(x-b).^2/c.^2-(y-d).^2./f^2)','inde',{'x' 'y'},'dep','z');

or, use a function handle:

g=fittype(@(a,b,c,d,f,x,y) a*exp(-(x-b).^2/c.^2-(y-d).^2./f^2),'ind', {'x' 'y'},'dep','z');