Semi-parametric estimation of covariate effects using the positive stable frailty model

Many biological and medical studies have as a response of interest the time to occurrence of some event, such as the occurrence of a particular symptom or disease, remission, relapse, death due to some specific disease, or simply death. In this paper we study the problem of assessing the effect of potential risk factors on the outcome event of interest through a parametric or semi-parametric frailty model where the lifetimes have a reason to be considered dependent. This dependence may arise because of multiple endpoints within the same individual or because, when studying a single endpoint, there are natural groupings between study subjects. The objective of this paper is to extend both parametric and semi-parametric approaches to regression analysis in which the lifetimes of individuals in a group are effected by the same random frailty which follows a positive stable distribution. Some comparisons of the properties of this frailty distribution with other frailty distributions are made and an example which assesses the effect of a treatment in a litter-matched tumorigenesis study is presented.     

Detailed nuclear structure studies far from stability

State-of-the-art spectroscopy of nuclei far from stability has achieved an extraordinary level of sophistication and detail in the last ten years. In principle, if a state can be populated, it can be characterized by its energy, spin, parity, and major decay paths. Sometimes its lifetime can be measured. In practice, one is confronted with enormous complexity. To convert raw spectroscopic data into nuclear structure data involves a complex process of disentangling gamma rays and conversion electrons into decay schemes. Specifically, coincidence techniques, especially coincidence intensities, play a crucial role in this process. Recent examples and methods from work done at UNISOR are presented.     

Discontinuous travelling wave solutions for certain hyperbolic systems

In an earlier paper on a malignant cell invasion model (Marchantet al., SIAM J. Appl. Math, 60, 2000) we introduced a novelform of discontinuous travelling wave solution. These solutionscould be studied easily by combining behaviour within a phaseplane with the Rankine–Hugoniot shock conditions, whichdescribe properties (such as the ratio of the jump discontinuitiesto the speed of propagation) that solutions may possess. Theseresults were new for several reasons. The shock conditions relateto hyperbolic equations (which the model is) but were appliedin a travelling wave ordinary differential equation phase planeusing techniques that usually apply to parabolic reaction–diffusionsystems. In addition the solutions possess singular behaviournear several points in the phase plane but in spite of thisthere exists a robust and stable family of physically interestingsolutions. In this paper we discuss two previously studied models, oneof detonation theory and one of angiogenesis. We show that eachof these models also possesses a family of discontinuous travellingwave solutions which was not previously discovered. Of particularinterest is the solution which has a blunt interface at thefront of the invading profile. In all three models it is thissolution that is seen to stably evolve from physically relevantinitial data, and for physically relevant parameter values. This work confirms the robustness of these novel travellingwave solutions and their applicability to a wider range of mathematicalmodelling situations.