Is a Random Number Generator Truly Uniformly Distributed?

AI Thread Summary
The discussion centers on the concept of uniformly distributed random numbers, specifically questioning whether such numbers, when drawn from the interval [0, 1], are irrational with probability 1. It is noted that under the standard Lebesgue measure, the set of irrational numbers has measure 1, while the set of rational numbers has measure 0, thus supporting the claim. Various methods for generating random numbers are shared, including a mathematical approach using squares of four-digit numbers and a practical method involving selecting digits from a telephone book. Participants emphasize the importance of understanding random variables and suggest resources for further study in probability theory. The conversation highlights the complexities of defining and proving properties of random numbers in mathematical contexts.
adityatatu
Messages
15
Reaction score
0
Can somebody prove the following:

A Random number which (truly) is uniformly distributed on [0 1] (or in fact any continuous random variable) is irrational with probability 1.
 
Last edited:
Physics news on Phys.org
In order for that problem to be "well-defined", you have to specify the measure. Assuming that the measure is the standard Lebesque measure, the set of all irrational numbers between 0 and 1 has measure 1 and the set of all rational numbers has measure 0.
 
I don" know the prove! but I know an interesting way to get random numbers.
just take a 4 digit number. get its square. keep two first digits away. then get the 4 next number az the next step iteration.
some initial values dosent work. can you tell them?!

for example:
1-0.3265
2-3265
3-3265^2=10660225
4-10_6602_25
5-6602 ----0.6602
6_ go to step 1.
 
The cardinality of the rationals is aleph_0 while that of the irrationals is aleph_1, I think !
 
It is easy to prove that the measure of the rationals is 0, because of countability. Make a list of rational numbers. Cover the nth member of the list (symmetrically) by an interval of length x/2n. All the rational numbers are then covered by a set of measure less than x, since the union of these intervals has measure less than the sum of the individual measures. Since x can be made arbitrarily small, the measure of the rationals is 0.
 
Another way to get random numbers:
Get ur telephone book, open it anywhere, choose a column, and pick the LAST digit in every phone number from the column…
 
adityatatu said:
Can somebody prove the following:

A Random number which (truly) is uniformly distributed on [0 1] (or in fact any continuous random variable) is irrational with probability 1.

I have a question:
What does random number exactly mean?
I have this question since I hear the word!
can you help me?
Thanks in advanced. :smile:
 
You need to first understand the concept of a random variable. It is a variable which has a value determined on the basis of a probability distribution. A random number is a random variable with a probability distribution uniform between 0 and 1.
 
dear mathman;
can you explain more or give me a reference to study?
thanks in advanced.
:smile:
 
  • #10
I am sorry to say that I haven't looked at any recent material on the subject. Feller's 2 volume "An Introduction to Probability ..." is a very good, but old, text. You can try probability theory with google.
 
  • #12
The wolfram reference is a good description for those using random numbers. However, you need to look at a good basic probability textbook to understand the mathematical foundations behind the concept of random variable.
 
  • #13
Thanks a lot for the informations
 
Back
Top