Statalist


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

st: RE: new package hangroot


From   "Nick Cox" <n.j.cox@durham.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: RE: new package hangroot
Date   Mon, 5 Nov 2007 16:27:59 -0000

This is welcome. Rootograms, in various flavours, not necessarily 
this one, go back at least to Tukey circa 1965. 
There are some references at [R] spikeplot, or in its references. 

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

Maarten buis

Thanks to Kit Baum a new package is available on -ssc-: -hangroot-. It
creates a hanging rootogram (Tukey 1977, Wainer 1974) comparing an
emprical distribution to the best fitting normal distribution. (Though
the article by Wainer was published earlier it cites a pre-release
version of (Tukey 1977) as its source.) 

A hanging rootogram is an alternative for a histogram with the normal
option. It differs from a histogram in two ways:

1) Instead of showing the freqencies it shows the square root of the
frequencies.  This way deviations from normality in the tails, where
the counts are small, become more visible.

2) The spikes or bars "hang" from the normal distribution instead of
"standing" on the x-axis. The deviations from normality are now shown
as deviations from a horzontal line (y=0) instead of deviations from a
curve (the normal curve). This makes it easier to spot patterns in the
deviations from normality.

I suspect that the primary usefulness of -hangroot- will be as one of
the last steps in graphically assessing the quality of a -regress-
model, by checking the normality of the residuals. However, I never
underestimate the creativity of users in finding new uses for existing
software. 

-hangroot- can be installed by typing -ssc install hangroot-.

-- Maarten

Referencces

Tukey, John W. (1977), "Exploratory Data Analysis", Addison-Wesley.

Wainer, Howard. (1974), "The Suspended Rootogram and Other Visual
Displays:  An Empirical Validation".  The American Statistician, 28(4),
pp. 143-145.

*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   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