News for hackers. Stuff that matters.
Re:Judge: Drinking tea, shopping at garden store probable ca... by echo_oddly on Monday December 28, 2015 @09:39PM (#10804848)
We can show with a few assumptions that the police thought that the prior probability of the leaves being marijuana (before the field test) is at least 41%.

A = leaves in trash are marijuana B = field test result is positive

First use Bayes' to get P(A|B) = P(B|A)P(A) / [P(A)P(B|A)+P(-A)P(B|-A)]

Assume the false positive rate, P(B|-A) = 0.7. Assume the true positive rate, P(B|A) = 1.

The result is P(A) = 0.7P(A|B)/(1-0.3P(A|B))

Assume to get a warrant the police need P(A|B)>0.5. This puts P(A)=0.41. However, probable cause is more strict. Assume we need P(A|B)>0.90. This puts P(A)=0.86.

So if you go to a garden store and you drink tea, the police are 86% sure there are illegal substances in your home.

There are no comments on this comment yet.