Aspects of modeling and application of survival-type data
Fallah, Lida
Fallah, Lida
Loading...
Publication Date
2018-02
Type
Thesis
Downloads
Citation
Abstract
Survival analysis is collection of methods for analyzing data where the outcome of interest is the time to an event and some of the observations are censored. Survival data can arise naturally from studies on, machines' time to break down (also known as reliability) to agricultural experiments on how some environmental conditions affect flowering, to medical cohort studies following-up cancer patients' survival and their reaction to treatments, etc. In industrial applications there are obvious benefits of progressive censoring (briefly speaking, removing live individuals progressively over time according to a censoring plan) in machine testing where effort, resource, and cost can be saved by early censoring. Furthermore, in agricultural applications, such as the serious threat of certain pests to sugar cane during the planting season or the maturation phase of the cane, biological control assays are used to study the survival of pests under exposure to pesticides. In addition, in recent decades, detecting the associations between patients gene expression profiles and phenotypic data is of increasing interest to aid in improving diagnosis and prognosis of patients and in facilitating treatment discoveries. To appreciate different aspects of survival data and its applications, this thesis puts together different methods for modeling such data and deals with the unique difficulties that each type of data brings to bear on data analysis.
Publisher
NUI Galway
Publisher DOI
Rights
Attribution-NonCommercial-NoDerivs 3.0 Ireland