A Test for Randomness

[Bayes]

New article: A Test for Randomness.

Creative system designers might get some new ideas from this. 

[Drazen]

Thanks Bayes

I signed second time as you instructed, but again no notification mail. Unless it comes with a bit of delay?

Cheers

[Bayes]

Hi Drazen,

Hopefully you'll get the notification tomorrow. The RSS file gets checked at 4 am GMT daily and then the emails are sent.

[Turner]

I keep thinking of something.

A random number cant be tested on its own. It can only be tested in a sequence of other random numbers

......a sequence of other random numbers that have absolutely no relation to each other

And....

past random numbers have no effect on future numbers, but future numbers arnt random unless they have them.

[Priyanka]

but future numbers arnt random unless they have them

I don't intend to start a debate. But that goes to my theory and the way I see the game Future and present always impacts the past

[Turner]

BTW...Bayes

Ive always wanted a good description of Z score....so thanks

[Bayes]

I think there's a lot of confusion in discussions because there is no agreed definition of what "random" means. For me it's quite simple, it just means lack of information.

link:://wmbriggs.com/post/2227/

[Turner]

For me it's quite simple, it just means lack of information.

But that lack of information cant be proven until you use past spins

Its probably a silly argument....but I am just saying.

[Bayes]

But that lack of information cant be proven until you use past spins

But how do you "prove" lack of information? I think I know what you're saying, but you're assuming that looking at past spins is a no no because they're independent. But that's just an assumption - how do you know it? - because they're "random" numbers!

It's a circular argument.

[PeaBea65]

I don't intend to start a debate. But that goes to my theory and the way I see the game Future and present always impacts the past

Without a doubt. 

[Turner]

It's a circular argument.

Of course

But is there such a thing as a random number or is it a number produced randomly?

I mean, one of these numbers is off a roulette wheel and one I just wrote down

6 and 32

Which is the random number. Are they both random because with my number, I only chose 1?

Is neither random because it can only pass a test in amongst other random numbers.

Im trying not to be facetious

[Priyanka]

Im trying not to be facetious

Sometimes the only way one can make sense is by being facetious. Random is one such subject

[Bayes]

But is there such a thing as a random number or is it a number produced randomly?

I would say it has to be produced by something. A random number is generated by some kind of mechanism or process which produces numbers which can't be predicted with any reliability because not enough is known about the generating process, or it is too complex for us to comprehend. E.g. radioactive decay. So it comes back to lack of information.

I mean, one of these numbers is off a roulette wheel and one I just wrote down

6 and 32

Which is the random number. Are they both random because with my number, I only chose 1?

They're both random for me, but only one is random for you.

[Toby]

Hi bayes,
Is your calculation box(at your site article) for even money bets?

We have got m(hits), n(fails), ans r(data ammount) to calculate the zcore

[Bayes]

Hi Toby,

No, I think you've misunderstood the purpose of this z-score. Any z-score just expresses the deviation from the average or expectation in terms of the number of standard deviations from it. What you need is the z-score for wins and losses, not the number of runs. And no, it's not just for even money bets although the final sequence is two-valued. There's no calculator for that ATM but I'll be adding one to the stats page soon, for those who don't like doing maths (99% of us  ).