calculate the mean square error

Given two arrays of double how do I calculate the mean square error

Answer 1

[Adam]

Calculate the error, square it and calculate the mean!

mean( ( a - b ).^2 );

[Guido Pastore]

but is it necessary that the two arrays have the same size ??

[Adam]

Yes, otherwise what are you calculating errors between? The errors are per point - you need a value and a value to compare it with for each point. If you don't have that then you can either throw away points that don't have a matching value or potentially interpolate values to compare them with, but that depends very much on the situation - 'inventing' data is not a great idea in many situations.

[Guido Pastore]

How can I interpolate values ??

I'll explain my situation to you.

I have two arrays.

1.The first array comes from a data acquisition performed through Kinect which has a higher acquisition frequency, in fact the size of this array is 1054x1.

2.The second array comes from a data acquisition performed through an Intel RealSense camera that has a lower acquisition frequency, in fact the size of this array is 601x1.

How can I solve the problem and then proceed to the calculation of the mean square error??

[Adam]

doc interp1

should allow you to resample the 601 up to 1054 (or the 1054 down to 601 if you prefer) and then you can compare per-point, bearing in mind that you are no longer comparing raw data.