R help - Oct 2006 - set.seed() and .Random.number

Hi R-users
I have two conditions. For each condition, 100 sets of 10 random numbers 
from N(0,1) need to be generated.  Here is my question.

At the begining I specify a seed number. I want to make the 100th set of the 
first condition and 1st set of the second conditon the same. What do I need 
to do ?

After generating 99th set of 10 random numbers and then saving .Random.seed 
then using .Random.seed for generating 100th set of the first condition and 
1st set of the second condtion. Is this right?

.Random.seed is a vector with 626 numbers, but set.seed() only accepts a 
integer number.
What do I need to do for saving .Random.seed and using the saved 
.Random.seed for setting set.seed() ?

Thanks

Taka

_________________________________________________________________
Get FREE company branded e-mail accounts and business Web site from 
Microsoft Office Live

Taka Matzmoto wrote:
> Hi R-users
> I have two conditions. For each condition, 100 sets of 10 random numbers 
> from N(0,1) need to be generated.  Here is my question.
> 
> At the begining I specify a seed number. I want to make the 100th set of
the
> first condition and 1st set of the second conditon the same. What do I need
> to do ?
> 
> After generating 99th set of 10 random numbers and then saving .Random.seed
> then using .Random.seed for generating 100th set of the first condition and
> 1st set of the second condtion. Is this right?
Yes, that's right.
> 
> .Random.seed is a vector with 626 numbers, but set.seed() only accepts a 
> integer number.
> What do I need to do for saving .Random.seed and using the saved 
> .Random.seed for setting set.seed() ?
As Prof Ripley pointed to me in a post a while ago out, the help 
?.Random.seed specifies that .Random.seed is the PRNG state, not the 
seed itself. So assign .Random.seed to a variable, and then assign the 
variable to it later:

 > runif(3)
[1] 0.919502 0.361622 0.847657
 > r <- .Random.seed
 > runif(3)
[1] 0.3898361 0.4727398 0.5698874
 > .Random.seed <- r
 > runif(3)
[1] 0.3898361 0.4727398 0.5698874


Cheers,
Gad


-- 
Gad Abraham
Department of Mathematics and Statistics
The University of Melbourne
Parkville 3010, Victoria, Australia
email: g.abraham at ms.unimelb.edu.au
web: http://www.ms.unimelb.edu.au/~gabraham