I have a question about setting the stream/state/seed of the random number generator.
I am essentially running a bunch 100+ Monte-Carlo simulations on r2010a. I want each simulation to produce different/independent (or nearly independent) sequences of random numbers. Ideally, I would be able to save the stream/state/seed so I could do it again. There is a lot of information, including some helpful posts on Loren's blog and related posts on the newsgroup, but I just want to make sure I am doing this correctly
From what I can tell I cannot use Mersenne Twister ('mt19937ar') since it does not support multiple streams. The Combined Multiple Recursive ('mrg32k3a') does support multiple independent streams. What I propose to do is generate a bunch of independent substreams (x) and save them to a .mat file:
>> N = 100; % Number of simulation runs
>> t = round(A*sum(clock));
>> s = RandStream('mrg32k3a');
>> s.Substream = t;
>> x = randi(B, N);
>> save('streams.mat', 'x');
Each of my simulations has an integer id (1 < id < N) which I will use to index into the list of "streams". Specially, in each simulation:
>> s = RandStream('mrg32k3a');
>> s.Substream = x(id)
Does this work?
How big do the scale factors A and B need to be?
I figured the size of A depends on the accuracy of the clock (Matlab used to suggest A=100 with the old algorithm). This seems to limit the streams to millisecond accuracy on the clock. Is there an advantage/disadvantage of going larger (maybe 1e3 or 1e4)?
What about B? Do I use the maximum integer size (2^51) or something else?
Any advice would be appreciated
On 12/13/2010 6:01 AM, Daniel wrote:
>>> N = 100; % Number of simulation runs
>>> t = round(A*sum(clock));
>>> s = RandStream('mrg32k3a');
>>> s.Substream = t;
>>> x = randi(B, N);
>>> save('streams.mat', 'x');
> Each of my simulations has an integer id (1 < id < N) which I will use
> to index into the list of "streams". Specially, in each simulation:
>>> s = RandStream('mrg32k3a');
>>> s.Substream = x(id)
> Does this work?
It does, but it may be more than what you need to do. There's no reason
that I can think of that simulations 1 through N can't just use
substreams 1 through N.
It seems like you're trying to inject some "extra randomness" into the
process by using clock and then drawing random integers. That most
likely is not necessary. If you're trying to get a "truely random"
result, then you ought to think about what that really means. The
typical MC simulation involves taking some kind of average ove many MC
iterations. I would argue that you should be using enough MC iterations
that the result of your simulation does not depend strongly on what
specific random numbers you use (assuming of course that they have
"good" distributional properties).
If you're goal is to be able to run the set of simulations and get
different results each time, without having to worry a lot about the
mechanics of making sure the random number stream is initialized
differently, then using clock can work nicely. But it seems like you've
described wanting repeatability. I would think that
s = RandStream('mrg32k3a');
s.Substream = id;
is all you need. Another possibility, using multiple independent
streams rather than substreams within one stream, would be
s = RandStream.create('mrg32k3a','NumStreams',1:n,'StreamInd',id);
Six of one, half a dozen of the other. The latter code allows for
(much) more than 2^76 random values for each simulation, but that's a
lot of random values in either case.
Hope this helps.
You can think of your watch list as threads that you have bookmarked.
You can add tags, authors, threads, and even search results to your watch list. This way you can easily keep track of topics that you're interested in. To view your watch list, click on the "My Newsreader" link.
To add items to your watch list, click the "add to watch list" link at the bottom of any page.
To add search criteria to your watch list, search for the desired term in the search box. Click on the "Add this search to my watch list" link on the search results page.
You can also add a tag to your watch list by searching for the tag with the directive "tag:tag_name" where tag_name is the name of the tag you would like to watch.
To add an author to your watch list, go to the author's profile page and click on the "Add this author to my watch list" link at the top of the page. You can also add an author to your watch list by going to a thread that the author has posted to and clicking on the "Add this author to my watch list" link. You will be notified whenever the author makes a post.
To add a thread to your watch list, go to the thread page and click the "Add this thread to my watch list" link at the top of the page.
No tags are associated with this thread.
A tag is like a keyword or category label associated with each thread. Tags make it easier for you to find threads of interest.
Anyone can tag a thread. Tags are public and visible to everyone.
The newsgroups are a worldwide forum that is open to everyone. Newsgroups are used to discuss a huge range of topics, make announcements, and trade files.
Discussions are threaded, or grouped in a way that allows you to read a posted message and all of its replies in chronological order. This makes it easy to follow the thread of the conversation, and to see what’s already been said before you post your own reply or make a new posting.
Newsgroup content is distributed by servers hosted by various organizations on the Internet. Messages are exchanged and managed using open-standard protocols. No single entity “owns” the newsgroups.
There are thousands of newsgroups, each addressing a single topic or area of interest. The MATLAB Central Newsreader posts and displays messages in the comp.soft-sys.matlab newsgroup.
You can use the integrated newsreader at the MATLAB Central website to read and post messages in this newsgroup. MATLAB Central is hosted by MathWorks.
Messages posted through the MATLAB Central Newsreader are seen by everyone using the newsgroups, regardless of how they access the newsgroups. There are several advantages to using MATLAB Central.
Your MATLAB Central account is tied to your MathWorks Account for easy access.
Use the Email Address of Your Choice
The MATLAB Central Newsreader allows you to define an alternative email address as your posting address, avoiding clutter in your primary mailbox and reducing spam.
Most newsgroup spam is filtered out by the MATLAB Central Newsreader.
Messages can be tagged with a relevant label by any signed-in user. Tags can be used as keywords to find particular files of interest, or as a way to categorize your bookmarked postings. You may choose to allow others to view your tags, and you can view or search others’ tags as well as those of the community at large. Tagging provides a way to see both the big trends and the smaller, more obscure ideas and applications.
Setting up watch lists allows you to be notified of updates made to postings selected by author, thread, or any search variable. Your watch list notifications can be sent by email (daily digest or immediate), displayed in My Newsreader, or sent via RSS feed.
You can also select a location from the following list: