x*****8
发帖数: 98 | 1 Suppose someone does 6 independent hypothesis tests using a .3 level of
significance. On each test, though he does not know it, the null hypothesis
is true. What is the
probability that he makes Type I errors on more than half the tests? What is
the probability that
he makes at least one Type II error?
谢谢了 | k*******a
发帖数: 772 | 2 第一个就是Binomial(6, 0.3) distribution
所以 more than half type I error = sum(dbinom(4:6, 6, .3)) =0.07
第二个，既然null is true, 怎么会有type II error 呢？ | x*****8
发帖数: 98 | 3 谢谢
【在 k*******a 的大作中提到】
: 第一个就是Binomial(6, 0.3) distribution
: 所以 more than half type I error = sum(dbinom(4:6, 6, .3)) =0.07
: 第二个，既然null is true, 怎么会有type II error 呢？
|
|
