Confidence interval
The range within which the true answer probably lies, given that you only measured a sample rather than everything.
Any measurement taken on a limited number of cases carries uncertainty. The confidence interval puts a number on it: instead of saying "the effect is minus 0.4 percent", it says "the effect is somewhere between minus 1.5 and plus 0.7 percent".
The practical rule is simple and unforgiving. If that range contains zero, you have not shown that any effect exists at all. The measured average might just be the noise of a small sample.
This is why a result quoted without its interval should be treated as decoration rather than evidence.
Across the nineteen iPhone launches since 2007, Apple’s share lost 0.4 percent on average on announcement day. But the interval runs from minus 1.5 to plus 0.7 percent. It contains zero, so the honest conclusion is that no effect has been demonstrated.
See also
This definition is part of the EPTA5 lexicon, written so that a reader who does not work in finance can follow our market notes without stopping. It is general information and not investment advice. Read the notes.