Name for binning a variable with respect to another variable(s)

[Ahmes]

Hi,

Sorry if it's not the right place as this question is a bit generic.

Let say we have N datapoint triplets x_i, y_i, z_i.
If we ask "how many points" there are in the interval X < x < X+ΔX and Y < y < Y+ΔY (where X and Y are some gridpoints) -- then this is a 2D histogram.

But in each such 2D bin there is some distribution of z values. How would you call a 2D map of the average (or whatever distribution moment) of the z values in the 2D bin around (X,Y)? I just can't think of the right term! It's not what I would usually consider interpolation.

I'll be very happy if someone knows the name of this procedure.

 

[Evalesco]

Hi, this is a great question! You could be looking for something called binning or histogram binning. This is when you divide a dataset into multiple bins and calculate the average or other statistic of the data in each bin. In your case, you are trying to calculate the average of the z values in the 2D bin around (X,Y). I hope this helps!