No. Let's say you've done 1000 flights and had 0 failures. Clearly the average failure rate is 0, but you may just have been lucky- there is a big error bar on the average failure rate. For example, if the real failure rate was worse than you'd hoped, say, 1 in 900 then the probability that you'd had no failure 1000x in a row = ( 1-1/900)^1000 = 32% that you have had no failures yet. So you'd need to do many more trials to prove that the failure rate was as good as you say with 95% accuracy.--- Randall Clague <[EMAIL PROTECTED]> wrote:You need 2998 flights to demonstrate that with 95% confidence.Eh? 1000 runs where X does not fail, therefore P(X_fail) < 1/1000. Ditto Y. These just happen to be the same 1000 runs.
In fact, after 1000 trials with no failures, the 95% confidence is a failure rate of only 1/330.
--
-Ian "Everything I say is a lie"
Motto: "You're Not Authorized to Know Our Motto."
"The future isn't what it used to be, but then it never was."
"Everbody loves my Baby, but my Baby loves no one but me."
"Predestination was doomed from the start."
"Everything takes longer than you expect, even when you take
into account Hofstadter's Law" -Hostadter's Law
