model fitting using lsqnonlin

y1=f(a,b,c,d,e,f,x,y,z) y2=f(a,b,c,g,h,i,x,y,z)

y1, y2 are known outputs. x,y,z are known inputs. a,b,c,d,e,f,g,h,i are the variables which I wish to find using lsqnonlin and lsqcurvefit. As there are two functions and variables are common, how shall I write fun in lsqnonlin(@fun,x0)? Thanks in advance

Answer 1

[Torsten]

Concatenate y1 and y2 to y=[y1 y2] and x1 and x2 to x=[x1 x2] and proceed as usual.

Best wishes

Torsten.

[Monika K]

Thanks for reply. I made mistake while asking que.. input x is itself matrix [x y z].

[Torsten]

In lsqnonlin, you could calculate the residuals as

for i=1:length(y1)
F(i)=y1(i)-f(a,b,c,d,e,f,x(i),y(i),z(i));
end
for i=1:lenghth(y2)
F(length(y1)+i)=y2(i)-f(a,b,c,g,h,i,x(i),y(i),z(i));
end

Best wishes

Torsten.