3 ms·
It might be obvious to you already, but saying it anyway just in case: every single explanation you get here, however nice of a story it sounds like to you, is
by kqr 4y ago
It might be obvious to you already, but saying it anyway just in case: every single explanation you get here, however nice of a story it sounds like to you, is pure speculation.
If a HN contributor actually could explain market movements, i.e. really grokked the causal network underlying equity prices, then they would make millions exploiting the ignorance of everyone else, and probably not sharing their secrets with the rest of the world.
- hansvm 4y agoIs that necessarily true? Is it not possible to use the extra information you have with hindsight to construct correct causal narratives and simultaneously also not be able to use such techniques to predict the future because you wouldn't have the requisite information?
- kqr 4y agoThis is a difficult epistemological argument that it took me some time to accept, but: how can you confirm the explanation is correct, if not by testing it as a hypothesis, i.e. using it to predict unknown (future) outcomes? Put differently, if you say "X causes Y, but only in this one instance we're discussing now and not in the future" then does X really cause Y, or was it a coincidence that they happened at the same time this once? The razor leads us to conclude that what's indistinguishable from coincidence is coincidence. ---- Edit: Oh, I see what you're saying! Something like "X causes Y but knowledge about X is only available after Y is already known"? That is possible, and it's even testable by blinding the person to Y while revealing X. Is it useful knowledge? Maybe!