Test For Bioequivalence Based On Generalized P Values

Nearly30years, The linear mixed-effect models is widely used, especially, in biomedicine,economic, financial, environmental science, sampling survey and engineering technologyengineering and so on. With the development of computer technology and the correspond-ing statistics software such as the SAS, SPSS, the application of linear mixed model is moreand more widely. About the linear mixed effects model, we have derived more perfectlytheory and good results of parameter estimation. This article mainly study the bioequiva-lence using generalized model P value method.Bioequivalence is the important problem to pharmacokinetics, the problem has some-thing with statistics. Bioequivalence test used to evaluate the test medicine (T) and regis-tered medicine (R) of bioequivalence to determine the effect quite, in new drug develop-ment and new Drug evaluation process plays an important role. Currently three typesof bioequivalence are distinguished: average, population and individual bioequivalence.Average and population bioequivalence can be assessed in two-period (non-replicated)crossover studies, whereas individual bioequivalence requires three-or four-period repli-cated studies, with a preference for four-period studies. The average bioequivalence onlyobserve the average level of the evaluation index, and don’t consider the variation amongthe individual; And population bioequivalence takes into account both the average level,and considering the variation between individuals;The individual bioequivalence exceptconsider average and individual variation outside, still considered the individual and druginteractions between.The research history and background of bioequivalence are introduced, then the con-cepts of generalized p value and generalized confidence intervals and regions are told inchapter1. Some statistical methods about average bioequivalence are introduced in chap-ter2. We use the generalized p value method to study the construction of confidence in-tervals and regions for population bioequivalence and individual bioequivalence. responsemean in generalized linear models with missing response in chapter3,4. Also, a class ofestimators for the parameters of interest is constructed, and the asymptotic distributionsof the proposed estimators are obtained. A simulation study indicates that the proposedmethods are comparable in terms of coverage probabilities and average lengths/areas ofconfidence intervals/regions. An example of a real data set is used for illustrating ourmethods.Our approach makes full use of the data information to improve the accuracy ofthe confidence regions. A simulation study explains the finite sample properties for pro-posed methods.