Skip to content

Question

7 messages · Davia Cox, Sean Davis, Uwe Ligges +2 more

#
Hello,

I have a problem that I am trying to solve and I am not sure how to  
do it in R.

Suppose, that 16 numbers are choosen at random from 0 to 9, what's  
the probability that their average will be between 4 and 6. I typed  
the following code:

set.seed(100)
sample(0:9, 16, replace =TRUE)
    [1] 3 2 5 0 4 4 8 3 5 1 6 8 2 3 7 6

Is what I got, however I realize the set.seed function locks in the  
number I get every time.
My question is in order to run a true random sample, wouldn't I have  
to use the runif function? And then deliminate the sample to show the  
numbers that lie between 4 and 6? If that's the case, how do I do that?


Davia S. Cox
517-575-8031 cell
davia.cox at gmail.com

"Human potential, though not always apparent, is there waiting to be  
discovered and invited forth." -William W. Purkey
#
Please disregard this message and don't post it to the web. I found  
the answer.
Thanks

Davia S. Cox
517-575-8031 cell
davia.cox at gmail.com

"Human potential, though not always apparent, is there waiting to be  
discovered and invited forth." -William W. Purkey
On Dec 13, 2005, at 6:20 AM, Davia Cox wrote:

            
#
On 12/13/05 6:20 AM, "Davia Cox" <davia.cox at gmail.com> wrote:

            
Just don't use set.seed() before every run (unless you want to always get
the same answers).  Set.seed() is available to allow you to generate
reproducible results, so not using it means that you will get a different
set of random numbers every time you run your "sample" from above.

Sean
#
Davia Cox wrote:

            
Yes, that's what set.seed is intended to do... otherwise don't use 
set.seed (and make your work unreproducible).
We have to disappoint you: Your computer cannot generate "true random 
samples".


 > wouldn't I have
No, if you want integers, the sample above fits perfectly well.
Is this a homework question?

Uwe Ligges
#
On 13-Dec-05 Uwe Ligges wrote:
At least on Linux (and probably most Unix) systems,
/dev/random must be a pretty good approximation (provided
you don't over-work it ... ).

Oh for the days of "shot noise"!

Best wishes,
Ted.


--------------------------------------------------------------------
E-Mail: (Ted Harding) <Ted.Harding at nessie.mcc.ac.uk>
Fax-to-email: +44 (0)870 094 0861
Date: 13-Dec-05                                       Time: 14:16:59
------------------------------ XFMail ------------------------------
#
On Tue, 13 Dec 2005 Ted.Harding at nessie.mcc.ac.uk wrote:

            
See

library(accuracy)
?trueRandom

(for some definition of 'true').
#
On 13-Dec-05 Prof Brian Ripley wrote:
Spot on! Thanks for this -- I wasn't aware of it previously.

The URL to HotBits in ?trueRandom is worth following, and the
descriptions in "How it Works":

  http://www.fourmilab.ch/hotbits/how.html

are entertaining reading.

Best wishes,
Ted.

--------------------------------------------------------------------
E-Mail: (Ted Harding) <Ted.Harding at nessie.mcc.ac.uk>
Fax-to-email: +44 (0)870 094 0861
Date: 13-Dec-05                                       Time: 16:47:37
------------------------------ XFMail ------------------------------