Following up on my post in the last open thread, I'm reading Understanding Uncertainty which I think is excellent.
I would like to ask for help with one thing, however.
The book is in lay terms, and tries to be as non-technical as possible, so I've not been able to find an answer to my question online that hasn't assumed my having more knowledge than I do.
Can anyone give me a real life example of a series of results, where the assumption of exchange ability holds and it isn't a Bernoulli series?
Let's say that we have a box of weighted coins. Some are more likely to fall heads; others tails. We pull one out and flip it many times. The flips are identical, so we can switch the order. They are independent conditional on knowing which coin was chosen, but ahead of time they are dependent, the one telling us about the choice of coin and thus about the other. De Finetti's theorem says that all exchangeable sequences take this form.
Added: Actually, de Finetti's theorem only applies to infinite sequences. Here's an example of a finite exchangeable sequen...
If it's worth saying, but not worth its own post (even in Discussion), then it goes here.