STAT-3090 Chapter Notes - Chapter 10.2: Type I And Type Ii Errors, Test Statistic, Sampling Distribution
Document Summary
Type i error and type ii error: if the value of the test statistic falls in the rr, we reject h. We conclude that the sample evidence is insufficient to reject h. Ha: drug safe: type i error: reject h0 when, in fact, it is true. (cid:862)drug is safe (cid:449)hen, in fact, drug is unsafe(cid:863) Consequences: not approving a safe drug does not give people a new medical treatment option. However it does not immediately puts people at the very high level of health risk: type i error, type ii error: Example 1: errors (cid:862)drug is safe (cid:449)hen, in fact, drug is unsafe(cid:863) (cid:862)drug is unsafe (cid:449)hen, in fact, drug is safe(cid:863) Probability of type i error should be small = level of significance should be very small. However, the smaller we make , the more likely we increase type ii error . The jury trial of an accused murderer is analogous to the statistical hypothesis process.
