|
7/3 |
2009/11/4-17 [Transportation/Car] UID:53496 Activity:nil |
11/4 I posted the probability question the other day. It was nothing as complicated as MTBF or the like. My situation is very similar to looking at auto accident rates, asking if the rate of caraccidents for any given driver is N%/year, what is the chance that for a given day that driver will have a crash? N% or N/365% or something else? Thanks again. \_ keywords: stats statistics \_ http://www.businessinsider.com/answers-to-15-more-google-interview-questions-that-will-make-you-feel-stupid-2009-11#if-the-probability-of-observing-a-car-in-30-minutes-on-a-highway-is-095-what-is-the-probability-of-observing-a-car-in-10-minutes-assuming-constant-default-probability-2 \_ You'd have to know something about the distribution of car crashes. Presumably they're more likely to happen around holidays and such. Few real-world situations have flat distributions. -tom \_ Agreed. However, if you're willing to assume a flat distribution, your probability should be (1-N%)^(1/365) neglicting leap years. -- yuen \_ That doesn't seem to make intuitive sense. If N=10% then your calculation says there is a 99.97% probability you WILL crash on any given day. \_ Look at the inverse. \_ Oops. I forgot the "1 -" part. It should be 1 - (1-N%)^(1/365). -- yuen \_ why are you giving away the fish? Teach a man to FISH!!! |
7/3 |
|