BANOVA {bear}R Documentation

Statistical analysis (ANOVA(lm), 90CI...)

Description

With a two-treatment, two-period, two-sequence randomized crossover design, ANOVA statistical model includes factors of the following sources: sequence, subjects nested in sequences, period and treatment. ANOVA will be applied to obtain estimates for the adjusted differences between treatment means and the standard error associated with these differences. Log-transformed BA measures will also be analyzed.

We investigate the assessment of equivalence in intra-subject variabilities of bioavailability between formulations. Point and interval estimates for the inter-subject and intra-subject variabilities are also provided.

Moreover, a normal distribution test procedure based on Spearman's rank and Pearson's correlation coefficient and Pitman-Morgan's adjusted F test are provided.

Finally, several tests for assumptions using the inter-subject and intra-subject residuals are discussed. Statistical tests for detection of outlying subjects such as Hotelling T2 are presented.

References

1. Chow SC and Liu JP. Design and analysis of bioavailability- bioequivalence studies. Chapman & Hall/CRC, New York (2009). 2. Liu JP and Weng CS. Detection of outlying data in bioavailability- bioequivalence studies. Statist. Med. 10, 1375-1389 (1991).


[Package bear version 2.3.0 Index]