2 ms·
> Two statistical dependent variables "can be made" independent, for example by feeding one into an PRNG No statistical test can assure statistical independenc
by naasking 7y ago
> Two statistical dependent variables "can be made" independent, for example by feeding one into an PRNG
No statistical test can assure statistical independence for all possible cases. The very fact that you pipe it through a PRNG means the output is deterministically correlated with the input, because PRNGs are deterministic, and some statistical test will be able to detect it. At the base level, a test that tries every conceivable PRNG, for instance.
- raverbashing 7y agoThat's why I said it's a theological problem In practice the universe can't know all the ways you can mix-up a variable
- naasking 7y agoWho says the universe can't "know" all the ways? That's like saying momentum can't be conserved because the universe can't know all the ways that we could transform momentum into other forms of energy and back again.
- raverbashing 7y agoMomentum comes from the interaction of forces, there's nothing that's "needed" to know. Two billiard balls colliding and a billiard ball colliding with a ball of mud conserves momentum (in a vacuum, in space, etc) the same because it all boils down to forces. F=dP/dt That's very different from "regardless how you throw the dice it will always "know" what you did".
- naasking 7y agoIn other words, we invented some concepts (force, momentum, energy) that exhibit a symmetry under various transformations. It's only "very different" in superdeterminism because that which is conserved doesn't yet have a widely accepted formulation that you've internalized the way you've internalized the other common concepts in physics.