Statalist The Stata Listserver


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

st: RE: graphing question


From   "Nick Cox" <n.j.cox@durham.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: RE: graphing question
Date   Wed, 5 Apr 2006 21:13:41 +0100

Setting aside the issue of multiplicity raised
by the numerous simultaneous -- and mutually
dependent -- tests possibly implied here, 

you can use -separate- to produce two distinct
response variables from one, and then employ
distinct symbols. 

It would still be a propaganda graph, to use
Tukey's term. (In revenge against poor referencers, 
I'll just give Tukey 1972 as the reference.) 

It might be more subtle to use (say) a line to indicate
threshold levels and let the user think about
the graph without the propaganda. Is the 
difference between 0.05% and 0.04% scientifically
interesting as well?  

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

Rajesh Tharyan
 
> Say I have a table which looks like this the *** shows the 
> values which are
> significant. Of course I can plot the percentages against 
> time. Is there
> someway to highlight the significant values in the 
> graph...Say for example
> by changing the size of the symbols only for the significant values.
> 
> Time	percentage	
> 
> -80	-0.02% 	
> -70	-0.01%	
> -60	-0.04%	
> -50	-0.05%	
> -40	-0.05%	
> -30	-0.07%***
> -20	-0.07%***
> -10	-0.16%***
> -5	-0.19%***
> -3	-0.17%***
> -2	-0.21%***
> -1	-0.27%***
>  0	 0.09%***
>  1	 0.22%***
>  2	 0.09%***
>  3	 0.08%***
>  5	 0.05%***
>  10	 0.04%	
>  20	 0.02%	
>  30	-0.01%	
>  40   -0.03%	
>  50	 0.02%	

*
*   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