Skip to content

Testing non-exhaustive Null and Alternative Hypothesis

2 messages · Michael Grant, Duncan Murdoch

#
On 07/09/2011 3:46 PM, Michael Grant wrote:
According to the standard hypothesis testing theory, that test is 
identical to the alternative of mu > 5, so t.test would be fine (if it's 
okay at all, you don't say distributional assumptions).  If you want to 
use a different basis for the test, you need to say what it is.

Duncan Murdoch