Use this URL to cite or link to this record in EThOS:
Title: Likelihood-based methods for the construction of cross-sectional age related standards
Author: Wade, Angela Mary
Awarding Body: University of London
Current Institution: University College London (University of London)
Date of Award: 1997
Availability of Full Text:
Access from EThOS:
Full text unavailable from EThOS. Please try the link below.
Access from Institution:
The problems associated with the construction of age-related standards have been recognised for almost half a century. The limiting factor has tended to be the difficulties associated with complex model fitting. Increased software capabilities have lead to increased interest over recent years. In chapter 1 both the past and current literature is reviewed. Chapters 2,3 and 4 use a likelihood based approach to create age related centiles. A continuous outcome is considered in chapters 2 and 3. Separate parametric curves are used to characterise changes in the skew, spread and median. These curves are then combined to create centiles and can be used to calculate z-scores throughout the age range. In chapter 4 similar methodology is applied to create standards for an ordinal outcome. Proportional odds models are used with parametric forms to model the logit of the probabilities. For both the continuous and ordinal outcomes curve forms were chosen that would asymptote at adult values. The likelihood approach has several advantages. Firstly, the likelihoods form a basis for significance testing between models. Secondly, profile likelihoods can be used to create confidence intervals for any of the model parameters, the centiles or z-scores obtained from them. Thirdly, if measurements are not independent then the correlation structure can be explicitly modelled and incorporated into the likelihood. This latter point is illustrated in chapter 3. The final chapter is an overview which discusses the relevance of the results given in chapters 2-4 and highlights potential areas for future development.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available
Keywords: Medicine