Go back
how to correct wikipedia edit error?

how to correct wikipedia edit error?

Science

h

Joined
06 Mar 12
Moves
642
Clock
16 Feb 15
3 edits
Vote Up
Vote Down

Originally posted by humy
I have now carefully edited all the changes in:

http://en.wikipedia.org/wiki/Probability_density_function

All the changes I made are in the "Example" section only.
Does everyone approve?
The main change there I made is replace where it previously said:

"Instead we might ask: What is the probability that the bacterium dies between 5 hours and 5.01 ho ...[text shortened]... illion nanosecond in a second, is (2 hour−1)×(1 nanosecond) = 2 × 10−9 / 3600 ≃ 5.556×10−13...."
I have just noticed that someone that calls himself Sbyrnes321 has reedited (not undo ) and enhanced some of my changes I made there to make them even better -good for him!

I knew my verbal skills are not as good as some peoples and I appreciate it when someone with obviously much better verbal skills than I helps me out with the wording.

He made it shorter and more to the point (close to halving the number of words! ) -something I find hard to do because one of my deficiencies is I always find it extremely difficult to verbally summarize things -don't know why. It certainly isn't due to lack of effort or motivation. Perhaps my (mild ) dyslexia is to blame for that?

R
Standard memberRemoved

Joined
10 Dec 06
Moves
8528
Clock
17 Feb 15
Vote Up
Vote Down

Originally posted by humy
On an unrelated maths problem (although still to do with probability ) :

Is there a way of simplifying the RHS expression in the equation:

y = ( x ^ n ) * ( ( 1 – x ) ^ ( e – n ) )

?

( don't know if this has any relevance but, for the application I have in mind, both n and e are natural numbers while x is a continuous random variable in the [0, 1] ...[text shortened]... ssible probabilities! )

I am trying to make it a bit easier for myself to finds its integral.
You might have to find a numerical approach for each n & e... there doesn't seem to be any worthwhile rearrangements to aid in the integration of the function in general.

h

Joined
06 Mar 12
Moves
642
Clock
17 Feb 15
5 edits
Vote Up
Vote Down

Originally posted by joe shmo
You might have to find a numerical approach for each n & e... there doesn't seem to be any worthwhile rearrangements to aid in the integration of the function in general.
Thanks for that.

With both e and n being natural numbers, I think I have found its integral which I think is:

definite integral of ( x ^ n ) * ( ( 1 – x ) ^ ( e – n ) ) dx from x=0 to x=1

= ( n! * ( e - n )! ) / ( e + 1 )!

I hope that is right. Can anyone confirm?

This would be fine for small/modest n and small/modest e but, for very large n or e, because it is too difficult to compute massively large factorials (my calculator can only find the factorial for whole numbers less that 70 ), I guess this formula would be impractical and I would have to use a numerical approach -unless someone can give me a formula for an approximation of the integral from x=0 to x=1 for very large n or e but avoids the factorial function?
Anyone?

R
Standard memberRemoved

Joined
10 Dec 06
Moves
8528
Clock
17 Feb 15
2 edits
Vote Up
Vote Down

Originally posted by humy
Thanks for that.

With both e and n being natural numbers, I think I have found its integral which I think is:

integral of ( x ^ n ) * ( ( 1 – x ) ^ ( e – n ) ) dx from x=0 to x=1

= ( n! * ( e - n )! ) / ( e + 1 )!

I hope that is right. Can anyone confirm?

This would be fine for small/modest n and small/modest e but, for very large n or e, b ...[text shortened]... e formula for an approximation of the integral from x=0 to x=1 for very large n or e ?
Anyone?
How did you come to find your result for the integral?

As far as I can tell, no elementary function has ( x ^ n ) * ( ( 1 – x ) ^ ( e – n ) ) as its derivative?

To check the result, what would help is the function in "x" before the integration (x=0 to 1) you performed?

Intuitively, I have to say the result is incorrect.

h

Joined
06 Mar 12
Moves
642
Clock
17 Feb 15
1 edit
Vote Up
Vote Down

Originally posted by joe shmo
How did you come to find your result for the integral?

I kind of 'cheated' by searching the net and finding the integration to a formula with the same structure although the letters used are all different:

http://math.stackexchange.com/questions/122296/how-to-evaluate-this-integral-relating-to-binomial

in this case, for the particular application I have in mind, I am only interested in the area under the curve from x=0 to x=1

R
Standard memberRemoved

Joined
10 Dec 06
Moves
8528
Clock
17 Feb 15
1 edit
Vote Up
Vote Down

Originally posted by humy
I kind of 'cheated' by searching the net and finding the integration to a formula with the same structure although the letters used are all different:

http://math.stackexchange.com/questions/122296/how-to-evaluate-this-integral-relating-to-binomial

in this case, for the particular application I have in mind, I am only interested in the area under the curve from x=0 to x=1
Ok...

I guess I don't have good intuitions about non elementary functions. (Apparently its the Beta Function)

h

Joined
06 Mar 12
Moves
642
Clock
18 Feb 15
5 edits
Vote Up
Vote Down

Originally posted by joe shmo
(Apparently its the Beta Function)
Thanks for telling me that. I just looked that up at:
http://en.wikipedia.org/wiki/Beta_function
and now I must gradually mull over it so I can understand it properly.
But, unless I am missing something, it doesn't appear to me to be exactly the same function but rather a similar function.
I haven't yet worked out how it relates to my formula if it relates at all!

looks like I may have to also study:
http://en.wikipedia.org/wiki/Beta_distribution

Cookies help us deliver our Services. By using our Services or clicking I agree, you agree to our use of cookies. Learn More.