|
I'm interested in building histograms over the values of data observed one-at-a-time in a sequence. However, the only techniques I know of for building histograms require ALL of the data to be present at once. This seems like it must be a common problem, so i'm assuming there are some algorithms out there that i'm unaware of. I'm looking for any pointers or suggestions to some of these. Btw- i'm happy to simplify and pick a # bins in advance if that makes things easier. |