Help understanding a normalization

Hi, in fact Ive been reading an article, they are working with velocity and acceleration:
So a number of observations (vt; at) (speed between 5 and 25 and acceleration between -2,2)
They said:
If we normalize these observations by the number of observations that fall into that speed bucket we get a two-dimensional (discrete) distribution on the rectangle R = (5; 20]*[-2; 2].

And then They did a v-a heatmap

I did not get what they intend by that


Active Member
In general, loosely speaking, normalization is transforming a variable so that its values are not too big and not too small. As such, they do not cause numerical roundoff or overflow errors; also they are amenable to plotting.

In the case of this paper, ask yourself: are the authors more knowledgeable in the field or less knowledgeable in the field than you?... Probably, heaps more. So they took their time to describe exactly and accurately what their procedure is but you said: no, no, no, those pages are not necessary. Why is that, my friend?

The only accurate way of analyzing an article is reading it (almost) fully and few people here would do it just like that. However, I would not be surprised if the authors performed a kind of division that constrained the transformed values to be in interval [a, b].