interval 0:5a and 1:5a . If the nominal level is a = 0.05,
the empirical Type I error rate should be between 0.025
and 0.075. Correspondingly, a procedure is considered to
be non-robust if, for any particular condition, its Type I
error rate is not within this interval. We chose this
criterion since it was widely used by many robust statistic
researchers (e.g. [6]; [12]; [16]; [17]) to judge robustness.
Nevertheless, for Guo and Luh (2000), if the empirical
Type I error rates do not exceed the 0.075 level, the
procedure can be considered robust. The best procedure is
the one that can produce Type I error rate closest to the
nominal (significance) level.
interval 0:5a and 1:5a . If the nominal level is a = 0.05,the empirical Type I error rate should be between 0.025and 0.075. Correspondingly, a procedure is considered tobe non-robust if, for any particular condition, its Type Ierror rate is not within this interval. We chose thiscriterion since it was widely used by many robust statisticresearchers (e.g. [6]; [12]; [16]; [17]) to judge robustness.Nevertheless, for Guo and Luh (2000), if the empiricalType I error rates do not exceed the 0.075 level, theprocedure can be considered robust. The best procedure isthe one that can produce Type I error rate closest to thenominal (significance) level.
การแปล กรุณารอสักครู่..
