How to compare two data sets, and make a general statement about how close they are from each other if we treat them as curves?

I want to compare two sets of data .. let's say I have n points of (x1,y1) data set ... and I want to compare it with m points of (x2, y2) data set ... Note that n and m are not equal ... I want to describe how close the two sets from each other.

I have experience with using techniques like (the mean absolute percentage error) or the (the coefficient of determination). However, these techniques are meant to compare two sets of data that are supposed to be identical. In this problem its not to meant to have points in same positions, but it is about to show the same general device behaviour.

Now I am thinking of fitting a curve to one set to produce a model then compare the other with it. However, the will mix the fitting error with the actual comarision results.

any suggestion please?

[KSSV]

As you can see the points are lying almost ont he curve......go for interpolation interp1, spline. Do the interpolation of m points with n points..check these interpolated values with the existitng values. And then you can fine RMS error and corellation coeffiecient between them.

[Ali Kareem Abdulrazzaq]

Thanks a lot KSSV,

I found your answer very useful

[Xiaochun Zhao]

I have the same problem as you.

Can you share the solution, bro?

[Nick]

I have same question.

Answer 1

[Ali Kareem Abdulrazzaq]

Thanks KSSV your replay provides me good solution

Best regards