PDA

View Full Version : [SOLVED] Types of Data Distributions



nirvehex
12-24-2014, 10:27 AM
Hello All,

This may be more of a math question, than an Excel question, however I'm running some analytics in Excel so I figured this was the best place to ask.

I'm given various types of distributions and asked to group them into like "buckets" or "tiers". So for example, (and this will vary every time) if I'm given a list of sales data for 300+ locations, I'm trying to get them into groups.

Here's my formulas so far:

J4 = # of Locations: 377
J5 = # of Bins: =ROUNDUP(LOG(J4,2)+1,0)
J6 = Bin Size: = (J7-J8)/J5
J7 = Max of Data Distribution = max(sales data) = 663
J8 = Min of Data Distribution = min(sales data) = 176

My question is:

Does anyone know any other methods for estimating the number of bins? The methodology I'm using above is called Sturges' formula.

How would other methods fit into my variables above?


Thanks to All and let me know if I can clarify anything further.

Kenneth Hobs
12-30-2014, 06:50 AM
Right, that is beyond what we deal with here.

Some other methods you can look at are shown in: http://en.wikipedia.org/wiki/Histogram

nirvehex
12-30-2014, 06:51 AM
Thanks for the clarification. Figured I would ask!

Kenneth Hobs
12-30-2014, 07:57 AM
Another site for alternatives is: http://robjhyndman.com/papers/sturges.pdf

This one shows how to do both Rices and Sturgess: https://www.mne.psu.edu/me345/Lectures/Histograms.pdf

This link has an Excel file where you can use that method and 2 others. http://www.prostatservices.com/statistical-consulting/downloadable-statistical-tools