MaplePrimes - answers and comments on Question, Generate a random variable VALUE with normal distribution

>

alex_01 — Thu, 18 Feb 2010 19:00:59 Z

restart:
with(Statistics):
Sample(RandomVariable(Normal(0, 1)), 10);

   [-0.479341933636560024, 0.784173699733298979, -0.500486574831822351,

     -0.491050898933502511, 0.237828352925436416, -0.148240640484583186,

     -0.0707527186858661478, 1.06583429443452870, -1.44945211081540992,

     0.571779637328814316]

Thank-you but I looked at

plummie — Thu, 18 Feb 2010 19:15:21 Z

Thank-you but I looked at that and I need to put the values into a matrix so it didn't work.
I have decided to use with(RandomTools)

Generate(distribution(Normal(0,1)) it seems to work
I also have Generate(distribution(Normal(0,.5))+I*Generate(distribution(Normal(0,.5))

Matrix

Scott03 — Thu, 18 Feb 2010 19:44:34 Z

If you are wanting to put the values into a Matrix try doing this in one step with something like the following:

> Matrix(3, 3, RandomTools:-Generate(distribution(Normal(0, 1)), makeproc = true));

Scott
Application Developer
Maplesoft

Matrix

Robert Israel — Fri, 19 Feb 2010 03:15:30 Z

> with(Statistics):
  X:= Sample(Normal(0, 0.5), 4) + I*Sample(Normal(0, 0.5), 4);
  ArrayTools:-Reshape(X,2,2);

I have looked at doing it in

plummie — Sat, 20 Feb 2010 14:23:05 Z

I have looked at doing it in one step but my off diagonal values have a different normal distribution to the diagonals and the matrix has to be hermitian. One way I did find to do it didn't allow me to generate eigenvalues of a large n x n case so I'm now attempting to build the matrix from scratch by generating values of each distribution and inputting them in the matrix using a for loop.

Obviously if there is a simpler way to do it, then I would appreciate your help.

Try this

Scott03 — Mon, 22 Feb 2010 20:11:14 Z

How about using this call?

Matrix(3, 3, RandomTools:-Generate(distribution(Normal(0, 1)), makeproc = true), shape = hermitian);

This uses the shape option to make the matrix the correct shape that you are looking for.

Scott