The combining of information: Investigating and synthesizing what is possibly common in clinical observations or studies via likelihood.
Year: 2008
Type: dissertation
Abstract: The combining of information: Investigating and synthesizing what is possibly common in clinical observations or studies via likelihood. A thesis submitted by Keith ORourke of Worcester College towards a D.Phil. degree in the Department of Statistics, University of Oxford, Trinity Term, 2003. The thesis is to develop an analytical framework for a exible but rigorous model based investigation and synthesis of randomized clinical trials regardless of outcome measure, probability model assumed or published summary available. This involves the identi
cation of relevant statistical theory, the development and adaptation of necessary techniques and the application of these to a number of examples. A new strategy for the investigation and synthesis of RCTs regardless of outcome measure, probability model assumed or published summary available was developed to accomplish this. No such general strategy has been explicitly set out before. It provides a quite general method, and with adequate sample information, results in arguably correct and adequate techniques for the assumptions made. A new method of numerical integration was developed to incorporate exible random e¤ects models; an importance sampling approach was developed to obtain the needed observed summary likelihoods and a Monte Carlo based diagnostic to assess the adequacy of sample information was produced but remains to be further researched.
Author: Keith O’Rourke
Language: English
Subfield: Statistics, Probability and Uncertainty
Field: Decision Sciences
Domain: Social Sciences
Open Access status: green
OpenAlex ID: W2397193839