Random Number Table

What are random numbers?

A computer cannot produce true random numbers and produces only pseudo-random numbers. A computer can reproduce the same sequence of digits if we repeat exactly the same algorithm. Modern algorithms create the illusion of having a true random generator, but it’s only an illusion.

In Matlab, there are two common functions for generating pseudo-random numbers: rand and randn.

Built-in function rand is meant for uniformly distributed pseudo-random numbers. R = rand(N) returns an N-by-N matrix containing pseudo-random values drawn from a uniform distribution on the unit interval. rand(M, N) returns an M-by-N matrix.  rand(M, N, P,...) returns an M-by-N-by-P-by-... array.  rand with no arguments returns a scalar.  rand(size(A)) returns an array the same size as A.

Built-in function randn is meant for normally distributed pseudo-random numbers. R = randn(N) returns an N-by-N matrix containing pseudo-random values drawn from a normal distribution with mean zero and std. deviation one.  randn(M, N) returns an M-by-N matrix. randn(M, N, P,...) returns an M-by-N-by-P-by-... array.  randn with no arguments returns a scalar. randn(size(A)) returns an array the same size as A.

Random numbers table

A random number table is a listing of random numbers where we can choose the quantity of random numbers desired, the maximum and minimum values of numbers in the table, and whether or not duplicate numbers are allowed.

If we need to generate 100 uniform random numbers on the set 0 : 1 in Matlab, we can just type rand(100, 1).

If we need to generate 10 uniform integers on the set 1 : n in Matlab, we can do this 

% Choose your maximum integer, n
n = 15;
% Generate 10 integers, from 1 to n
f = ceil(n * rand(1, 10));
% Display the first five elements in your table
f(1 : 5)
 

If we have a specific group of numbers already defined in a table, and want to make a random selection of numbers in that table, we can generate a random index and then see what number of the table is the corresponding one.

For example, let’s say that we have a vector x that has values from -10 to 10 in 0.05-steps. We want to take three random elements from that table.

% Vector from -10 to 10, in 0.05-steps
x = -10 : 0.05 : 10; 

% Generate 3 random indices on the set 1 : length(x)
% Display those indices, just to keep them visible
n = length(x);
ix = ceil(n * rand(1, 3)) 

% Display our corresponding three x-values
x(ix)
 

We get from Matlab 

ix =    340   211    82
ans =   6.9500    0.5000   -5.9500
 

This means that our indices were randomly selected from all of the possible indices in x (401 possible integers, due to the size of the vector), and the selected values were within the [-10, 10] interval and were a multiple of 0.05, as per our specifications. Of course, if you repeat the experiment, you're going to get different indices and values.

Now, let’s generate a uniform distribution of random numbers on a specified interval [a, b]. To do this, multiply the output of rand by (b-a), then add a. For example, to generate a 4-by-3 array of uniformly distributed random numbers on the interval [20, 95], we can type

a = 20; b = 95;
x = a + (b - a) * rand(4, 3) 

and we could get 

x =
   35.5350   63.1361   43.4514
   65.5399   53.8569   20.9647
   67.2416   23.2921   48.7975
   47.7858   22.0389   71.2337