3. Frequentist versus Bayesian Statistics

  1. Guosheng Yin

Published Online: 9 JAN 2012

DOI: 10.1002/9781118183335.ch3

Clinical Trial Design: Bayesian and Frequentist Adaptive Methods

Clinical Trial Design: Bayesian and Frequentist Adaptive Methods

How to Cite

Yin, G. (2011) Frequentist versus Bayesian Statistics, in Clinical Trial Design: Bayesian and Frequentist Adaptive Methods, John Wiley & Sons, Inc., Hoboken, NJ, USA. doi: 10.1002/9781118183335.ch3

Publication History

  1. Published Online: 9 JAN 2012
  2. Published Print: 19 DEC 2011

ISBN Information

Print ISBN: 9780470581711

Online ISBN: 9781118183335



  • Bayesian approaches;
  • copula;
  • frequentist approach;
  • generalized estimating equation (GEE);
  • generalized method of moments (GMM);
  • hypothesis testing;
  • maximum likelihood method;
  • multivariate distribution


A multivariate distribution can be easily constructed by linking marginal distributions through a copula. The maximum likelihood method is the most widely used frequentist approach to estimation and inference. The generalized method of moments (GMM) is an estimation and inference procedure that has gained much popularity in econometrics. Hypothesis testing typically involves a null hypothesis H0 and an alternative hypothesis H1, and each of them poses a statement on the parameter of interest ?. In contrast to the marginal approach such as the generalized estimating equation (GEE), the random effects model is a subject-specific approach by explicitly formulating the dependence structure through random effects. This chapter provides a statistical background for clinical trial designs. It introduces the estimation and inference procedures separately for the frequentist and Bayesian approaches.

Controlled Vocabulary Terms

Bayesian probability; continuous multivariate distributions; copula; discrete multivariate distributions; hypothesis testing; maximum likelihood estimation; statistical probability