Re: Random Variable on rationals in [0, 1]



I have found F using first 200 rationals.
It wont be exact. But it should give pretty good
idea about graph of F as the probabilities which I ignored
would not sum up more than 2^(-200).

But this is not exact solution or some closed form solution.

---
Regards,
Sujit P Gujar
IISc Bangalore.
Web:http://clweb.csa.iisc.ernet.in/sujit

On Feb 12, 8:10 am, Sujit <sujit.gu...@xxxxxxxxx>
wrote:
Is my question clear?

Yes, your question is clear, but the answer is not.
To parphrase: you
construct a function f from the rationals in [0,1] to
the integers
{1,2,3,...} and assign probability 2^(-i) to the i'th
rational under
f. You then want the cumulative distribution F(x)
back in [0,1]. You
claim to have found F for the first 200 rationals
using Matlab, but I
don't believe it. Between x_i and x_j (1 <= i < j <
200) there will (I
suspect) be infinitely many other rationals that
occur very far out in
the list, and just looking at the first 200 rationals
won't pick up
all these intermediate points.

R.G. Vickson


---
Regards,
Sujit P Gujar
IISc Bangalore.
Web:http://clweb.csa.iisc.ernet.in/sujit

Hello

Let Omega = set of all rationals in [0, 1].
Now define random variable X on Omega.
As rationals are countable, we can order them.
We will define probability mass function(pmf) as
follows:
for i^{th} rational x,  f(X=x) = 2^(-i).

p/q be the representation of a rational st,
q != 0, gcd(p,q) = 1.
Now we can map each rational to integers with map
p/q |--> 2^p3^q

and for 1 ( p =1, q =1) map it to 2*3 and for 0
(p=0
q=1) to 3.

Now we order rationals according to the order
of their image under above map.
I mean is, we will say rational x is ith rational
in Omega if its image is i^{th} integer in wrt to
above map.
Now, Let F be the cumulative distribution
function of
X.
How to find out F(x) for x in [0,1]?
If we want F(x) to be correct up to some 2^{-i},
we
may consider
first i+1 rationals and find out which are less
than
or equal to x and sum their probabilities.

Is there any way to construct explicit formula to
find
F(x) in terms of x for the above random variable?
That is, put x in the formula and get F(x).

How in general the graph of F(x) will be? I guess
it
will be
continuous everywhere except rationals in [0,1].

(Using Matlab)I have considered first 200
rationals
(according the order defined above) and plotted
F.
I have attached the diagram for those who are
interested
in getting rough idea abt F.

---
Regards,
Sujit P Gujar
IISc Bangalore.
Web:http://clweb.csa.iisc.ernet.in/sujit



.



Relevant Pages

  • Re: Probability of picking a positive rational number at random
    ... There _are_ distributions on the rationals, ... Specifying such a distribution is equivalent ... probability questions such as the ones you are asking, or, if a closed ... So obviously this density function yields a finite result when asked ...
    (sci.math)
  • Re: Random Rational?
    ... |>number has an even denominator"? ... the result is plainly not literally a probability. ... of a set of rationals under the height ordering. ... absolute value that had an even denominator in lowest terms tends toward ...
    (sci.math)
  • Re: Factoring and rationals
    ... and _compute_ the empirical probability that the ... See the word "convention" in my original msg. ... there are several equally plausable enumerations to choose ... over rationals "chosen at random", and it looks messier to me too. ...
    (sci.crypt)
  • Re: Relative Cardinality
    ... >> with probability 0 it will be a rational. ... The set of rationals in is a measurable set with measure zero and ... the set of irrationals in ias a measurable set with measure 1. ... A uniform disribution on a countable set is not possible. ...
    (sci.math)
  • Re: forcing representation of denominator 1?
    ... call it PRINT-ODDS instead). ... The REPL printer is a debugger tool, ... I just found it _convenient_ to use rationals in this case. ... "given a probability between 0 and 1 of the event E, ...
    (comp.lang.lisp)