I understand the difference between linear curve fitting and interpolation. In interpolation, the targeted function should pass through all given data points whereas in linear curve fitting we find the general trend of dependent variable. The cost function could be the distance between them.
If we keep on going with same sense of cost function, are not in case of interpolation the difference between data points zero ?
Should we not call interpolation as non-linear curve fitting ?
If not then what is the difference
Best Answer
Not at all! Interpolation is
You can use many different methods for interpolation including linear interpolation and polynomial, or spline curves.
When you are fitting curve to the data it is up to you to decide how close do you want it to fit the data. Nothing stops you from choosing the curve that perfectly fits to your data.
We are talking about interpolation when you use the fitted curve to re-create, or guess, the unobserved datapoints.