Open Archive 6
[ Read Responses | Return to the Index ]
[ Previous | Previous in Thread | Next in Thread | Next ]
Tools of the trade
Posted by Cyrus on 14 November 1999, at 2:38 a.m., in response to Re: Methinks he is talking..., posted by Bootlegger on 13 November 1999, at 10:49 p.m.
"sitting at a nickel slot (I don't know why it has to be a nickel machine) and watching the roulette wheel.."
When necessary to go live on that slot (something like an attendant coming by for the 3rd time and giving you looks), you'll be paying for the clocking time less than if sitting at a dollar slot, simple as that.
"once you have found what you think is a "biased" wheel, how do you know whether or not all you have seen is nothing more than standard deviation in action?"
That's the toughest part of all, avoiding what statisticians call a Type-I error, i.e. seeing biases where in fact there's none. With enough measured data and the aid of some statistical tools, like the chi-square test as you know, we can reasonably ascertain whether we are witnessing randomness disguised or not. So, for starters, I will take the actual number of occurences for each roulette number and the expected number of occurences, say A and E respectively. Deduct E from A, square the result and divide it by E. Do that for every E and A and sum. That's the χ2 (or chi-square): the greater the value of chi-square, the higher the probability that my data has not been randomly generated.
The Open Archive 6 is maintained with WebBBS 2.24.060398.