|
Equivalence
and Non-inferiority Procedures included in version 3.0.
For non-inferiority- and equivalence hypotheses it's quite common to use confidence intervals
instead of one- and two one-sided tests, respectively. The confidence level is then chosen to
correspond to the significance level of the test. In the following situations the sample
size can be calculated using the confidence interval approach.
|
|
|
|
-
Equivalence,
Confidence Interval
- Diference
of Means, t-distribution
- Ratio
of Medians, Lognormal distribution
- Difference
of Proportions, Binomial distribution
- Difference
of Paired Proportions
- Comparison
of Survival, Exponential distribution
-
Bioequivalence,
Two-way Crossover Design
- Test
approach
- Confidence
interval approach
-
Non-inferiority,
Confidence Interval
- Diference
of Means, t-distribution
- Ratio
of Medians, Lognormal distribution
- Difference
of Proportions, Binomial distribution
- Difference
of Paired Proportions
- Comparison
of Survival, Exponential distribution
|
|