On 2/7/2007 8:20 AM, Marc Bernard wrote:
> Dear All,
>    
>   I am looking for an R function or any other reference to generate a series 
> of correlated Binomial (not a Bernoulli) data. The "bindata" library can do 
> this for the binary not the binomial case.

Ted asked how you want your series correlated.  Another question is how 
you want it "binomial":  do you want each value to be binomial with 
parameters conditional on previous values, or do you want each to be 
marginally binomial with some fixed parameters?

I can only think of two trivial solutions that meet both conditions 
simultaneously:  the i.i.d. sequence, and a sequence that repeats 0 
indefinitely.  I'd be interested in hearing if anyone knows of any 
non-trivial examples of sequences where the distributions are both 
conditionally and marginally binomial, or a proof that there are none.

Duncan Murdoch

______________________________________________
[email protected] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.

Reply via email to