[R] [PS] rmultinomial() function

[Mary Black]

Thanks Ben and Eric.  I've already tried rmultinom(), and there is a rmultinomial() function as well (which is in the multinomRob package). The rmultinomial() is supposed to be a random number generator for the multinomial distribution.  There is an argument "long" which if set to TRUE or FALSE, utlize different generators, but the documentation doesn't explain any further. Thus, my confusion about the output.  If anyone can explain further how the two generators are different, I would love to know!  



----- Original Message -----
From: Ben Fairbank <BEN at SSANET.COM>
Date: Wednesday, March 19, 2008 7:36 am
Subject: RE: [PS]  [R] rmultinomial() function
To: Mary Black <mhbridge at usc.edu>, r-help at r-project.org

> Mary --
> 
> The dmultinomial function (try ?Multinomial, noting that it is an 
> uppercase M) has a "log" option, which, if set to TRUE, returns 
> logarithms of
> probabilities, but that is for computing probabilities, not generating
> samples.  Perhaps the "long" you referred to is a misprint for "log?"
> 
> In any case, try ?Multinomial, and give rmultinom() another try.  
> Note,however, that its output gives the _number_ of each of the 
> sampled items
> produced in a single sample, not the sequence of draws.  If you 
> need the
> sequence, then I think that the reply from Erik Iverson tells you how
> best to proceed.
> 
> Ben 
> 
> -----Original Message-----
> From: r-help-bounces at r-project.org [mailto:r-help-bounces at r-
> project.org]On Behalf Of Mary Black
> Sent: Tuesday, March 18, 2008 5:29 PM
> To: r-help at r-project.org
> Subject: [PS] [R] rmultinomial() function
> 
> After scouring the online R resources and help pages, I still need
> clarification on the function rmultinomial().  I would like to 
> create a
> vector, say of 100 elements, where every element in the vector can 
> takeon the value of 0, 1 or 2, and where each of those values have a
> specific probability. ie. the probability a given element in the 
> vector= 0 is 0.06, 1 = 0.38, 2 = 0.56 (probabilities sum to 1). 
> Can I use
> rmultinomial() function to do this?  
> 
> The following code does not seem to produce the result I need, but 
> thissort of code is all I could find the R "help" pages:
> 
> > rmultinomial(100,c(0.06,0.38,0.56))
> [1]  3 29 68
> > rmultinomial(100,c(0.06,0.38,0.56),long=TRUE)
>  [1] 3 3 2 2 3 2 3 3 2 3 2 3 3 3 2 2 3 3 2 3 1 2 3 3 3 3 3 3 2 3 
> 3 2 3
> 3 3 2 3 2 2 3 2 3 2 3 3 3 2 1 3 3 1
> [52] 2 3 2 2 3 3 2 2 2 1 3 3 2 3 3 3 3 2 3 3 3 3 2 3 2 3 3 2 3 3 
> 2 3 3
> 2 3 2 3 3 2 3 3 3 2 3 2 2 2 2 2
> 
> Also, I don't really understand the difference between the default
> long=FALSE and long=TRUE.  The R "help" simply states that you use 
> "longTRUE to choose one generator, FALSE to choose another one"; 
> however I
> could not find any documentation that described what the difference
> between those generators is.  Any clarification would be greatly
> appreciated! 
> 
> Thanks,
> 
> Mary
> 
> ______________________________________________
> R-help at r-project.org 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.
>