Totally valid Bayesian reasoning.
That was bothering me too. But I'm ok with "then you must have been exposed to the flu, and because you didn't get a shot you're feeling it"
-
-
Yeah there's definitely some language shorthand going on here. Minimal examples I'm trying to construct don't have quite the same effect
-
A: Someone was hit by a trolley B: That's because you pulled the lever A: I didn't pull the lever B: Then it's bc you didn't pull the lever
-
A: The police showed up at my house. B: That's because you set your home alarm [and it had a false positive that called the police].
-
A: I didn't set my home alarm. B: Then it's because you didn't set it [and burglars weren't scared away so you're filing a police report].
-
A: I'm broke B: That's because you play the lottery A: I don't play the lottery B: That's why you're broke
-


Also works with B unchanged and A substituting "rich" instead of "broke". 
-
The key is that when you learn P(flu-shot) = 0, having observed sorethroat, you have to update your prior on P(exposed-to-flu) to very high,
-
And that update switches the relative positions of P(sore-throat|flu-shot) and P(sore-throat|~flu-shot).
- 2 more replies
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.