China
Africa
Explore chapters and articles related to this topic
Statistical Methods for Assessment of Biosimilars
Published in Wei Zhang, Fangrong Yan, Feng Chen, Shein-Chung Chow, Advanced Statistics in Regulatory Critical Clinical Initiatives, 2022
Yafei Zhang, Vivian Gu, Xiuyu Julie Cong, Shein-Chung Chow
To overcome the problem of pairwise comparison, Zheng et al. (2017) proposed a simultaneous confidence approach based on the fiducial inference theory as an alternative to the method of pairwise comparison for similarity assessment of the three arms (i.e., one test group and two reference groups).
Content Uniformity Testing
Published in Emmanuel Lesaffre, Gianluca Baio, Bruno Boulanger, Bayesian Methods in Pharmaceutical Research, 2020
Steven Novick, Buffy Hudson-Curtis
Regulated content uniformity testing is not commonly performed with Bayesian procedures. This may be due to the difficulty in implementing Bayesian methods without a statistician or pre-canned software. Still, there are some notable Bayesian or Bayesian-like procedures in the literature for testing content uniformity. Some of the authors use generalized pivotal quantity (GPQ) methods (Weerahandi, 2004), a frequentist method that may be used to calculate confidence intervals. Hannig (2009) demonstrated that GPQ analysis is a special case of fiducial inference, a frequentist framework that resembles Bayesian methods. This section gives a brief overview of some of the available resources for implementing Bayesian approaches. Though all of the Bayesian methods in this section use vague or non-informative priors, the authors of each work note that real information could be incorporated into the prior distribution. In the first part of this section, we will discuss Bayesian techniques that can be used to improve upon or augment frequentist approaches, particularly relevant to ASTM E2810, which assesses a batch relative to USP<905>, and the PTI-TOST approach. In the second part of this section we will discuss statistical assurance, a Bayesian technique that can be used to assess risk by examining the expected performance of a batch against a particular test. Statistical assurance may be seen as an extension of the operational characteristic curves used in the frequentist framework.
Generalized fiducial methods for testing the homogeneity of a three-sample problem with a mixture structure
Published in Journal of Applied Statistics, 2023
Pengcheng Ren, Guanfu Liu, Xiaolong Pu
Here we explain the general procedure of generalized fiducial inference. Suppose that we have the DGE et al. [15], of et al. [23]. Hence the generalized p-value for the one-sided hypothesis
Estimation of the probability content in a specified interval using fiducial approach
Published in Journal of Applied Statistics, 2021
Ngan Hoang-Nguyen-Thuy, K. Krishnamoorthy
In this article, we provide some simple solutions based on the fiducial approach to the aforementioned problems. The concepts of fiducial distribution and fiducial inference were introduced by Fisher [8,9]. Even though there are some severe criticisms concerning the interpretation of fiducial distribution (Zabell [27]) and not a popular statistical method, Efron [7] has noted in Section 8 of his paper that ‘maybe Fishers biggest blunder will become a big hit in the 21st century!’. The fiducial approach was resurfaced in the name of generalized variable approach introduced by Tsui and Weerahandi [24] and Weerahandi [25]. Hannig [11] has noted that the generalized variable approach is a special case of the fiducial approach, and all the results obtained using the generalized variable approach can be obtained using the fiducial approach. The fiducial approach is a useful tool to find solutions to many complex problems with satisfactory frequentist properties. See Clopper and Pearson [3], Garwood [10] and Chapman [1] for some classical results. For other problems where fiducial inference led to exact CIs, see Dawid and Stone [5], the articles by Krishnamoorthy and Mathew [16,17]. Applications of the fiducial approach to estimate the process capability indices (PCIs) can be found in Mathew et al. [22] and the recent article by Edirisinghe et al. [6].