1
$\begingroup$

For many hypothesis tests, Monte Carlo methods are used to estimate the empirical $p$-value which is defined as $$p=\#{(T_{sample} > T_{observed})}/N.$$ Is there a name for the samples with $T_{sample} > T_{observed}$? And for the samples with $T_{sample} < T_{observed}$?

$\endgroup$
1
  • $\begingroup$ "Samples having more/less extreme value of statistic"? I don't think there is any specific term. $\endgroup$
    – amoeba
    May 13, 2015 at 17:23

0

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.