[Math] estimating bucket size for distribution graphs

[Adam Hardy <ah...@cyberspaceroad.com>]

If I want to distribute my data into buckets for a distribution bar chart or 
graph, is there a way to use commons math to get a sensible bucket size for the 
distribution?

I'm thinking about some fraction of the standard deviation, or am I on the wrong 
trail there?

I'd appreciate any tips!

Thanks
Adam


---------------------------------------------------------------------
To unsubscribe, e-mail: user-unsubscribe@commons.apache.org
For additional commands, e-mail: user-help@commons.apache.org

Re: [Math] estimating bucket size for distribution graphs

[Phil Steitz <ph...@gmail.com>]

Adam Hardy wrote:
> If I want to distribute my data into buckets for a distribution bar
> chart or graph, is there a way to use commons math to get a sensible
> bucket size for the distribution?
> 
> I'm thinking about some fraction of the standard deviation, or am I on
> the wrong trail there?
> 
> I'd appreciate any tips!

The optimial bin size depends on the shape of the distribution and
the range of values represented in the data.  The
EmpiricalDistributionImpl class in the random package can be used to
 experiment with different bin sizes, computing bin counts and
descriptive statistics within bins.

Phil

> 
> Thanks
> Adam
> 
> 
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: user-unsubscribe@commons.apache.org
> For additional commands, e-mail: user-help@commons.apache.org
> 


---------------------------------------------------------------------
To unsubscribe, e-mail: user-unsubscribe@commons.apache.org
For additional commands, e-mail: user-help@commons.apache.org

Re: [Math] estimating bucket size for distribution graphs

[Ted Dunning <te...@gmail.com>]

I would think that the difference between the 75-th %-ile and the 25-th
%-ile would be more to the point.

On Thu, Jun 10, 2010 at 5:44 AM, Adam Hardy <
ahardy.struts@cyberspaceroad.com> wrote:

> If I want to distribute my data into buckets for a distribution bar chart
> or graph, is there a way to use commons math to get a sensible bucket size
> for the distribution?
>
> I'm thinking about some fraction of the standard deviation, or am I on the
> wrong trail there?
>
> I'd appreciate any tips!
>
> Thanks
> Adam
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: user-unsubscribe@commons.apache.org
> For additional commands, e-mail: user-help@commons.apache.org
>
>