A controversial application of Bayesian reasoning

Join the discussion
Ask a follow-up here, or get your own question answered by working scientists, mathematicians and engineers — people, not an autocomplete.
Real named experts · corrections over time · the nuance an AI answer skips
1 reply · 3K views
p78653
Messages
6
Reaction score
2
The gain in odds that aliens are visiting Earth (A) due to ##n## independent reports of close encounters (C) is given by:
$$\frac{\rm Odds(A|C)}{\rm Odds(A)}=\left[\frac{\rm Prob(C|A)}{\rm Prob(C|\bar A)}\right]^n.$$
Let us assume that we have good cases such that an alien explanation (##a##) is as likely as a non-alien explanation (##\bar a##):
$$\rm Prob(C\ \&\ a) = Prob(C\ \&\ \bar a).$$
Therefore
$$\rm \frac{Prob(C|A)}{Prob(C|\bar A)}=\frac{Prob(C\ \&\ \bar a)+Prob(C\ \&\ a)}{Prob(C\ \&\ \bar a)}=2.$$
Thus the gain in odds that aliens are visiting Earth is given by
$$\frac{\rm Odds(A|C)}{\rm Odds(A)}=2^n.$$
Accordingly we only need ##50## good cases of close encounters to raise the odds by a factor of ##10^{15}## which will overpower any reasonable prior bias against alien visitation.

Is this analysis correct?
 
  • Skeptical
Likes   Reactions: weirdoguy, Vanadium 50 and PeroK
Physics news on Phys.org
p78653 said:
Is this analysis correct?
No matter whether the application of Bayes' theorem is correct, the analysis is sensible only if we are allowed the assumption that "we have good cases such that an alien explanation (##a##) is as likely as a non-alien explanation (##\bar a##)" .

That assumption is inherently speculative so this thread violates the long-standing forum rule about UFO speculation and has been closed.
 
Last edited:
  • Like
  • Haha
Likes   Reactions: Agent Smith, Vanadium 50 and Dale