Statistical Modelling in Biostatistics and Bioinformatics

This book presents selected papers on statistical model development related mainly to the fields of Biostatistics and Bioinformatics. The coverage of the material falls squarely into the following categories: (a) Survival analysis and multivariate survival analysis, (b) Time series and longitudinal data analysis, (c) Statistical model development and (d) Applied statistical modelling. Innovations in statistical modelling are presented throughout each of the four areas, with some intriguing new ideas on hierarchical generalized non-linear models and on frailty models with structural dispersion, just to mention two examples. The contributors include distinguished international statisticians such as Philip Hougaard, John Hinde, Il Do Ha, Roger Payne and Alessandra Durio, among others, as well as promising newcomers. Some of the contributions have come from researchers working in the BIO-SI research programme on Biostatistics and Bioinformatics, centred on the Universities of Limerick andGalway in Ireland and funded by the Science Foundation Ireland under its Mathematics Initiative.

Presents new implications for statistical analyses of frailty survival models with structural dispersion Features new methods of model-based clustering in Bioinformatics as well as of detecting genetic sweeps in population genetics Provides new insights into non-proportional hazards survival models and the use of H-likelihood methods Presents results on the new class of HGLMs - hierarchical generalized non-linear models

Autorentext

Professor Gilbert MacKenzie has a research background in Epidemiology, Biostatistics and Mathematical Statistics. His current research interests include multivariate survival modelling, frailty modelling and covariance modelling. He has published a wide range of research papers and reports and is a past President of the Irish Statistical Association. He holds an adjunct Professorship in Statistics in the University of Limerick and was a visiting Professor in Statistics at ENSAI, France, from 2010 to 2011.

Professor Defen Peng was a visiting Professor in Statistics and senior Research Fellow in the BIO-SI research programme at Limerick from 2009 to 2010. Professor Peng originally worked in the field of Economics at Zhongnan University of Economics and Law, PRC. She has published widely and is currently pursuing several areas of Statistics, such as: survival analysis with frailty, bivariate survival analysis, and the stability of regression models with categoricalcovariates.

Inhalt
Preface.- An Appreciation - John Nelder, FRS.- Introduction.- Survival Modelling: Hougaard - Multivariate Interval-Censored Survival Data: Parametric, Semi-Parametric and Non-Parametric Models; MacKenzie and Ha - Multivariate Survival Models Based on the GTDL; Lynch and MacKenzie - Frailty Models with Structural Dispersion; Martinez and Hinde - Random Effects Ordinal Time Models for Grouped Toxicological Data from a Biological Control Assay.- Longitudinal Modelling & Time Series: Haywood and Randal - Modelling Seasonality and Structural Breaks: Visitors to NZ and 9/11; Allais and Bosco - Forecasting the Insolvency Risk of the Customers of an Automotive Financial Service; Xu and MacKenzie - On Joint Modelling of Constrained Mean and Covariance Structures in Longitudinal Data.- Statistical Model Development: Payne - Hierarchical Generalized Nonlinear Models; Durio and Isaia - Comparing Robust Regression Estimators to Detect Data Clusters: A Case Study; Coffey, Hinde and Garcia - FiniteMixture Model Clustering of SNP Data; Peng and MacKenzie - Discrepancy and Choice of Reference Subclass in Categorical Regression Models.- Applied Statistical Modelling: Ramsey - Statistical Methods for Detecting Selective Sweeps; Brophy, Gibson, Wayne and Connolly - A Mixture Model and Bootstrap Analysis to Assess Reproductive Allocation in Plants; Conde and MacKenzie - On Model Selection Algorithms in Multi-Dimensional Contingency Tables.- Postscript: Durio and MacKenzie - Obituary: Professor Ennio Isaia.