3 ms·
Multiple 3-sigma excesses explained by the same model are potentially much more significant from a Bayesian perspective than several unrelated 3-sigma excesses.
by jessriedel 5y ago
Multiple 3-sigma excesses explained by the same model are potentially much more significant from a Bayesian perspective than several unrelated 3-sigma excesses.
- exmadscientist 5y agoThat's very true, but the candidate LEP signal is pretty weak, given that they were explicitly on the hunt for this exact sort of thing. I agree with the parent poster: move along (unless you're directly working on this!), and then get happy news later if something interesting does emerge. Just don't wait up for it.
- jessriedel 5y ago> the candidate LEP signal is pretty weak, given that they were explicitly on the hunt for this exact sort of thing. Isn't the fact that they were explicitly hunting for it a reason to take it more seriously? In contrast, if this signal was discovered after trawling the huge dataset for a hundred possible hypotheses, that would be a reason to discount it, Bayes-wise.