MATLAB: How to do a multivariate nonlinear least squares problem using the LSQNONLIN function in the Optimization Toolbox 3.1.2 (R2007b)

Question

I wish to solve a multivariate nonlinear least squares problem using the LSQNONLIN function. I tried the example from the documentation for this but the the following commands appear to work only for 1 independent variable :

function F = myfun(x)
k = 1:10;
F = 2 + 2*k-exp(k*x(1))-exp(k*x(2));

Answer 1

In order to solve a multivariate non-linear least squares problem, you need to define input 'x' as a matrix, where each row corresponds to an

independent variable. However, since you can only pass a vector, you would

need to first reshape the matrix into a vector and reshape it back into a matrix

in the function.

Hence a modification of the code will be as follows:

function F = myfun(x)
% reshape x back to a matirx
k = 1:10;
F = 2 + 2*k-exp(k*x(1))-exp(k*x(2));