a Distribution Analysis formula.


by qpwimblik
Tags: distribution funcion
qpwimblik
qpwimblik is offline
#1
Mar7-12, 03:22 AM
P: 22
Using this formula the prime numbers should distribute themselves in a seemly chaotic balancing act between the negative and positive axis on your graph.
like this.

http://www.pyramidus.org/Images/Dist...onFunction.jpg

And the formula.

1/2*((-1+abs(prime(n))+abs(prime(1+n)))*(abs(prime(n))+abs(prime(1+n)))-(-1+abs(prime(2+n))+abs(prime(3+n)))*(abs(prime(2+n))+abs(prime(3+n)))-(-1+abs(prime(4+n))+abs(prime(5+n)))*(abs(prime(4+n))+abs(prime(5+n)))+(-1+abs(prime(6+n))+abs(prime(7+n)))*(abs(prime(6+n))+abs(prime(7+n)))-2*prime(1+n)+2*prime(3+n)+2*prime(5+n)-2*prime(7+n))
Phys.Org News Partner Science news on Phys.org
SensaBubble: It's a bubble, but not as we know it (w/ video)
The hemihelix: Scientists discover a new shape using rubber bands (w/ video)
Microbes provide insights into evolution of human language
qpwimblik
qpwimblik is offline
#2
Mar7-12, 05:48 AM
P: 22
I wonder if you could see if a number is a prime or not using analysis of numbers put into this formula while trying to make a sequence of primes. So by finding a sequence of primes you can see if your number is a part of the sequence. It probably wouldn't be a fast primality test for small numbers but who cares about small numbers say below 200000000. One thing I can see straight away is that there is nearly the same amount of primes below 0 as above 0 and From tests I have found that if one of the primes is even out by one nth prime the results are distorted and the graph becomes unbalanced. Feel free to play with the function yourself.

One thing I know the formula is useful for is for improving nth term approximations because the closer you can get your nth term formula in place of the prime function to the resemble the graph above the closer you are to understanding how to at least partially mimic the behaviour of the Primes which in-turn could help you to improve your pi(x) approximations. I know you've followed the crowd and looked at things with far bigger Big O's but 100 years ago you couldn't compute with as bigger number crunches as you can now with computers such that you can now test long equations quickly. Am I crazy.
qpwimblik
qpwimblik is offline
#3
Mar7-12, 06:38 PM
P: 22
Better yet.
From my studies of this formula you might be able to find Pn's ever further Pn's apart so long as the nth gaps in the sequence are all the same size so you could find Pn sequences with massive gaps. meaning that you might be able to find big Pi's say pi(10^600) very fast.

qpwimblik
qpwimblik is offline
#4
Mar7-12, 08:04 PM
P: 22

a Distribution Analysis formula.


The logic behind the formula

First we have have pattern 1
1,1,2,1,2,3,1,2,3,4,1,2,3,4,5 ... or A002260

we'll call this pattern A(n)

And we have pattern 2
1,2,1,3,2,1,4,3,2,1,5,4,3,2,1 ... or A004736

we'll call this pattern B(n)

Here's how we extrapolate n from A(n) and B(n).

n = ((A(n) + B(n) - 1) (A(n) + B(n))) / 2 - (B(n) - 1)

Now I replace A(n) and B(n) with Prime(n) and Prime(n+1) and see what output I get using the formula above.
Not satisfied with the usefulness of the results (I'm sure anyone who's even tried to think knows that moment)

Anyway yes so I decided to minus the formula from itself with a sequence of values of Pn to see how the distribution would fair with my best results coming from minusing the minused output from a minused output using a sequence of Pn's, what you see in the initial post is the Simplified formula.

sorry it's a bit complicated
qpwimblik
qpwimblik is offline
#5
Mar7-12, 09:57 PM
P: 22
Red = The main sequence of 8 Primes of Pn being dealt with using the main equation
Blue = The same sequence as Red but out by a small amount between 0 and +7 for each value in the sequence

note. keep a close eye on Reds activities at 0 and the Difference between Red and Blues lines just above zero.

http://www.pyramidus.org/Images/Dist...nOfIntrest.jpg

I think we should call my discovery the Wimblik Equlibrium
qpwimblik
qpwimblik is offline
#6
Mar8-12, 07:23 PM
P: 22
The formula for assessing 3 primes at a time is this.

1/2*(3*prime(n)^2-4*prime(1+n)^2-prime(n)*(3+2*prime(1+n))+(-3+prime(2+n))*prime(2+n)+2*prime(1+n)*(3+prime(2+n)))

And the formula to get very close to the lines the distribution makes is this with e use here being the number e.

lineNumber*0.98*(1-1/2*n*(1+n)-(2+e)*ln(n)+1/2*(-1+n+(2+e)*ln(n))*(n+(2+e)*ln(n)))

which all looks like this.

http://www.pyramidus.org/Images/OnTheLine.jpg
qpwimblik
qpwimblik is offline
#7
Mar10-12, 03:43 PM
P: 22
Using xlogx instead of Pn

http://www.pyramidus.org/Images/xLogx.jpg
qpwimblik
qpwimblik is offline
#8
Mar11-12, 02:05 PM
P: 22
A function iterating in on itself.

watch near zero.

http://www.pyramidus.org/Images/BigDistribution.jpg
qpwimblik
qpwimblik is offline
#9
Mar11-12, 04:36 PM
P: 22
An interesting signature look at the primes distributed.

http://www.pyramidus.org/Images/Complex.jpg
qpwimblik
qpwimblik is offline
#10
Mar12-12, 05:32 AM
P: 22
An attempt to guess the distribution with a random Integer variable in the formula.

http://www.pyramidus.org/Images/ComplexEq.jpg
qpwimblik
qpwimblik is offline
#11
Mar12-12, 06:50 AM
P: 22
My VIP find (Rudimentary pattern in the distribution of Primes)

http://www.pyramidus.org/Images/VIPDistribution.jpg

noted to myself must spend a year animating that.
qpwimblik
qpwimblik is offline
#12
Mar12-12, 07:55 AM
P: 22
Different Random number generation methods seem to exhibit different regularities here's one
example.

http://www.pyramidus.org/Images/NotRandom.jpg
Norwegian
Norwegian is offline
#13
Mar12-12, 09:21 AM
P: 144
Since I like the challenge of understanding other peoples thoughts and ideas, I have spent some minutes looking at the posts above.

So we start with a huge expression, and what I assume is its graph, for values of n up to 30000. It is not entirely clear to me what is meant by abs(prime(n)), though I am able to make a guess. Also, the expression can be considerably tidied up by introducing some short notation for repetitive sub-expressions.

I do not get much out of the next couple of posts, then, in the 4th, some explanation is finally revealed. Two sequences are given, and there is a formula for going from its values, and back to n. The two sequences are replaced by prime(n) and Prime(n+1), but the same formula is used to produce some number, and the big expression in the first post is a more involved version of this formula. I still wonder what the significance of this might possibly be.

Then on the next picture, you mention the Wimblik equiulibrium, without clearly stating what that is. This is followed by more graphs. I am sorry, but I think I have failed to understand this thread. On to the next.
qpwimblik
qpwimblik is offline
#14
Mar14-12, 12:46 PM
P: 22
I was just quickly probing Formulae to look for signature in the nth Prime distribution and I found it. So I know now that the distribution is not random just Infinitely dynamic which does mean that it's !Hard to even approximate which most think is so, anyway Hard does not equal impossible.
qpwimblik
qpwimblik is offline
#15
Oct23-12, 08:27 PM
P: 22
this is my Christmas Primes distribution analysis.
http://www.pyramidus.org/Images/ChristmasPrimes.gif


Register to reply

Related Discussions
Formula for the Electric Field Due to Continuous Charge Distribution Introductory Physics Homework 3
bayes formula and picking variance from distribution General Math 3
Why is my lamda different from the formula used by the book in Poisson Distribution? Calculus & Beyond Homework 0
Deriving the binomial distribution formula Calculus & Beyond Homework 1
barometric height distribution formula problem Advanced Physics Homework 1