Least squares fit/line fit for 3D data

I have 3D data that I'd like to get a least squares fit from. Once I have this fit with an equation, I'd like to transform new data with it...so I need the code and to understand where to plug the new data into whatever equation comes from it. Can anyone help? Much appreciated.

Thanks

[Matt J]

Duplicate of

https://www.mathworks.com/matlabcentral/answers/494646-3-d-line-of-best-fit-from-origin-to-cloud-of-data-points

Answer 1

[Star Strider]

For a linear regression, this is straightforward:

B = [x(:) y(:) ones(size(x(:)))] \ z(:);                    % Linear Parameters
z_fit = [x(:) y(:) ones(size(x(:)))] * B;                   % Fitted ‘z’

For a nonlinear regression, we would need sto see your model.

[Matt J]

This looks like a plane fit to me. A 3D line fit would result in 2 algebraic equations.

Also, the fit looks like it assumes no errors in x and y.