Hello, Everyone! :)
This one should sound easy but here it goes anyway:
What's the probability that AX^2 + BX + C = 0 has real roots if A, B, C are independent random variables with Uniform (0,1) distribution?
alain verghote - 09 May 2005 11:52 GMT
Dear Rodrigo,
Roots of trinom are real when discriminant >= 0
or B^2-4*A*C >= 0
You have to compute Prob(B^2-4*A*C >= 0) A,B,C
uniform on interval [0,1] ,
you may experiment the case with a uniform random func. :
(Rand1)^2-4*Rand2*Rand3 = , indices means castings
must be different ;follow the signs and values (to
estimate the mean );
to go further : study the Law of B^2 and A*C ,
Courage,Alain.
Ara M Jamboulian - 09 May 2005 19:00 GMT
> Hello, Everyone! :)
>
[quoted text clipped - 3 lines]
> real roots if A, B, C are independent random
> variables with Uniform (0,1) distribution?
Consider the horizontal a-c plane and the b-axis to be vertical. You need to calculate the volume of the solid above the surface b^2 = 4*a*c and below the plane b = 1.
Pat Ballew` - 10 May 2005 12:20 GMT
I tried doing a double integral of the cross sectional area parallel to the a-c plane (in terms of a) [c=1-(b^2/4) (1/a) from z^2 / 4 to 1
and then the height from 0 to 1 (in terms of b)
and then took 1- the result
and got just over .2544 ...
Trying the same thing with several thousand simulations using Fathom software gives essentially the same answer...
Pat Ballew
Lakenheath UK
matt271829-news@yahoo.co.uk - 12 May 2005 13:12 GMT
> I tried doing a double integral of the cross sectional area parallel to the a-c plane (in terms of a) [c=1-(b^2/4) (1/a) from z^2 / 4 to 1
> and then the height from 0 to 1 (in terms of b)
[quoted text clipped - 5 lines]
> Pat Ballew
> Lakenheath UK
I get an exact answer of 5/36 + Log(2)/6, or 0.2544134...
Rodrigo Viana Rocha - 28 May 2005 05:24 GMT
Thanks, everyone!
Weeks ago I was helpless about it and no one would answer it. I tried to solve using many ways and I finally found something before your answers pop up! I am amazing that people do care about PROBABILITY on this math forum. It seems there is no one specialized on it.
I have written a small .tex file on this matter and I would be glad to share it with anyone interested.
I thank you all for your right answers (at least I found the same results)...
Cheers,
Rodrigo