Re: [R] Association between discrete and continuous variable

[Spencer Graves]

     My recommendations would depend on the nature of the discrete and 
continuous variables.  What is the application?  Is one variable random 
and the other a control variable that might be used to predict?  If the 
continuous variable is random, what about using the R^2 produced by 
"lm"?  For the case where the discrete variable is random, I just tried 
"www.r-project.org" -> search -> "R site search" for "coefficient of 
determination for glm".  This produced 4 hits, the first of which (dated 
Mon Jul 21 2003) cited 3 papers published between 1991 and 1997. 

     Mutual information 
(http://www.engineering.usu.edu/classes/ece/7680/lecture2/node3.html), 
as you suggest, would be appropriate when you have appropriate models 
for the random nature of both variables, separately and together. 

     More generally, what problem are you trying to solve with this?  
Who will use the numbers?  How will they use them? 

     hope this helps.  spencer graves    

Murray Jorgensen wrote:
I'm wondering if mutual information al la Cover & Thomas (1991, Ch 2) 
is not the killer association measure for all types of random variables?

Murray Jorgensen
PS  Yes, this is probably OT!
Richard A. O'Keefe wrote:
What's the reommended way, in R, to determine the strength of
association between a discrete variable and a continuous variable?
Yes, I have read the manuals, trawled the archives, &c.
______________________________________________
[EMAIL PROTECTED] mailing list
https://www.stat.math.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! 
http://www.R-project.org/posting-guide.html



______________________________________________
[EMAIL PROTECTED] mailing list
https://www.stat.math.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html