Data probability/density

#1
Hi, hope im in the right place.

Im trying to find the best way to characterize the different data on this space? to 3 draw regions (or lines with uncertainty) that will describe the data with the same color.

I initially thought to take each data set (each color) and to make a plot of data probability (e.g. a region in which 90% of the data are inside it), but it will be biased toward the origin as a lot of data clustered there.

Any suggestions?
 

hlsmith

Omega Contributor
#5
Not my area, but I wonder if you could translate into exponential distribution or what base you would need to generate each line. BGM, has a good point, though I am not familar with finding or defining clusters our groups. Always seems like a completely different area of stats or language than I know!