# Generate Multivariate Normal Distribution

How do I generate a set of data of Multivariate Normal Distribution, give mean, std dev? Any available free software to do the job?
LVL 33
###### Who is Participating?
I wear a lot of hats...

"The solutions and answers provided on Experts Exchange have been extremely helpful to me over the last few years. I wear a lot of hats - Developer, Database Administrator, Help Desk, etc., so I know a lot of things but not a lot about one thing. Experts Exchange gives me answers from people who do know a lot about one thing, in a easy to use platform." -Todd S.

Commented:
There is VBA code to do exactly this at http://www.anthony-vba.kefra.com/vba/vba4.htm

Cheers

Dave
0
Author Commented:
Sorry. Choose to delete wrong question.
0
Commented:
It looks like the other question has already been deleted so this one (as the earlier question) is live

Regards

Dave
0
Commented:
0

Experts Exchange Solution brought to you by

Your issues matter to us.

Facing a tech roadblock? Get the help and guidance you need from experienced professionals who care. Ask your question anytime, anywhere, with no hassle.

Commented:
No probs :)
0
Author Commented:
Yes, but I need a multivariate one instead.
0
Commented:
Are you talking about multiple independent normally distributed variables or a group of variables with an associated covariance matrix?
0
Author Commented:
How do you interpret the below statement?

"To generate the discrete normal distribution, we ¿rst generated the mul-tivariate normal distribution, and then discretized the values of each attribute into equal-width ranges."
0
Commented:
That means to divide the interval spanned by the numbers into a number of subintervals each having same length. Then take your data and determine what interval each point belongs to. Finally, you count the number of occurances in each subinterval.

As a simple example, you might see something like below for 11 sub-intervals:
1: *
2: *
3: **
4: ****
5: ******
6: ********
7: *******
8: *****
9: ***
A: *
B: *

0
Author Commented:
Do you mean we will only generate ONE series of normal distribution numbers?
0
Commented:
Te answer to that goes back to the original statement:

"To generate the discrete normal distribution, we ¿rst generated the mul-tivariate normal distribution, and then discretized the values of each attribute into equal-width ranges" [key phrase "each attribute"]

One needs to know how many attributes are being generated. If one, then rkursem had the full answer, If two and a bi-variate normal distribution is desired, then ozo was correct. For more more then two you would have to define the relationship between the attributes and their distributions. A simple model would be to assume for N attributes an n-space matrix with each attribute applied orthogonaly to the rest. Then each attribute would be itself normaly distributed along its axis and you would generate n sets of normaly distributed numbers.
0
Author Commented:
How to split myself? I can't seem to find the option.
0
###### It's more than this solution.Get answers and train to solve all your tech problems - anytime, anywhere.Try it for free Edge Out The Competitionfor your dream job with proven skills and certifications.Get started today Stand Outas the employee with proven skills.Start learning today for free Move Your Career Forwardwith certification training in the latest technologies.Start your trial today
Math / Science

From novice to tech pro — start learning today.