Probability could this theory be exact>?

[eljose79]

Probability...could this theory be exact>?...

In fact let,s suppose we throw a dice...N times (with N tends to infinity)..then coul be a fixed point in the probability function..in this sense..for example after throwing dice N times (with big N we would always have the same number ,for example you throw the dice and after N times of throwing it you get the same number...is that possible?...can probability give exact results (or at least with little error) in measuring things?

 

[russ_watters]

No. Probablity is probability.

 

[Njorl]

I had a lot of trouble understanding what you were saying. Let me know if I am assuming wrongly. I am assuming this is what you meant:

Roll a die (I assume six-sided) N times. N is very large, approaching infinity if you wish. The same result comes up every time, assume that result is 3 for argument's sake. Does this mean that for a large enough N that we can say that 3 will always be the result?

Not quite. We can take the data - a large number of 3's and nothing else - and assign a confidence factor to a range of probabilities that the next result will be a 3. For 100%, the confidence is zero. If we include a range of probabilities, extending a tiny bit below 100%, the confidence increases. For an arbitrarily large N, we can have an arbitrarily small range of probabilities, from 99.99...9% to 100% and an arbitrarily high confidence. The larger N is, the higher the confidence factor is. It would be foolish to say there is only a 1/6 chance for a 3 to result. The data is evidence that the die is probably not true.

If you did not know what a coin was, and someone reported to you the results of a million coin flips as 499678 heads and 500322 tails, you could assign probability ranges to heads and tails, and quote confidence factors for those ranges. This will not rule out something other than heads or tails coming up. Remember, you don't know a coin has only 2 sides.

Njorl