I want to fit a curve to data, where the dependent and the independent variables both have errors. Usually Matlab assumes, that only the dependent variables have errors.
So my question is: Does Matlab offers methods to perform fits where both errors are taken into account?
My regards, Christian.
What you are referring to are a type of regression problems called error-in-variable models. They may go by other names depending on methods used.
MATLAB does not ship these methods as far as I know. However our friends at File Exchange has several submissions that may be exactly what you are looking for.
Here are few examples: