Ako nakupovať
Doručenie
Nákupný poradca
Bootstrap methods for multi-factorial dose-response designs
Angličtina
128 b
30 dní na vrátenie tovaru
Mohlo by vás tiež zaujímať
Bifactorial designs are used to test for the efficacy§ of fixed combinations of two (or more) drugs. Researchers are§ interested in the question if a combination has a§ significantly higher efficacy than either component drug§ alone. The power of the existing methods strongly depends on§ a nuisance parameter representing how much the respective§ means differ between the groups treated by the component§ drugs. Furthermore, the methods are only applicable under§ rather strict assumptions on the data. In this work, the§ distributions of the respective test statistics are§ approximated by bootstrap methods with data-based estimation§ of the marginal differences to overcome these limitations.§ This gives interesting insights about the role of unknown§ nuisance parameters in resampling procedures. The methods§ described are also expanded to related problems and more§ general questions and their power is evaluated by simulation§ studies. Finally, some examples for typical applications are§ worked out.
Informácie o knihe
Angličtina
Kategórie
