Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

RE: st: RE: bihistogram


From   "Nick Cox" <n.j.cox@durham.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   RE: st: RE: bihistogram
Date   Wed, 27 Aug 2008 15:39:01 +0100

Thanks for the plug. -stripplot- is on SSC. 

For comparison of two distributions I would add -qqplot- and -kdensity-.


Nick 
n.j.cox@durham.ac.uk 

Ronan Conroy

On 27 Aug 2008, at 12:16, Rajesh Tharyan wrote:

> Hi sorry about the post with no link. Not sure what happened
>
> http://www.itl.nist.gov/div898/handbook/eda/section3/bihistog.htm

Looking at this, I am still not sure that this is the best way of  
displaying the data. For several reasons

1. Histogram shape is notoriously sensitive to the width and starting  
position of bins.
2. The eye is better at comparing side-by-side plots where position on  
the same scale is used to make the comparision
3. A histogram does not show the data
4. A histogram cannot be used to infer descriptive statistics

I would recommend, as an alternative, Nick Cox's -stripplot- with - 
vertical- and -box- options. Suddenly you see far more of the data,  
and are in a far better position to make comparisons.

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index