Frailty models that yield proportional hazards
โ Scribed by Odd O. Aalen; Nils Lid Hjort
- Publisher
- Elsevier Science
- Year
- 2002
- Tongue
- English
- Weight
- 124 KB
- Volume
- 58
- Category
- Article
- ISSN
- 0167-7152
No coin nor oath required. For personal study only.
โฆ Synopsis
Proportional hazards is commonly assumed in survival analysis. This assumption appears to be mainly a mathematical convenience, made in order to interpret Cox models in a simple manner. There remains the questions of whether this assumption is actually true to reality in a su cient number of cases, and whether the assumption relates to other ways of modelling survival data. We shall look at the latter issue, in pointing out that certain frailty models may yield proportional hazards in a natural way. We consider frailty distributions determined by Lร evy processes. In our framework, the hazard ratio may be interpreted as the ratio between the squared coe cients of variation of the frailty distributions.
๐ SIMILAR VOLUMES
This paper presents a mixture model which combines features of the usual Cox proportional hazards model with those of a class of models, known as mixtures-of-experts. The resulting model is more flexible than the usual Cox model in the sense that the log hazard ratio is allowed to vary non-linearly
Long-term excess hazards for cancer survival sometimes tend to zero or become negative even though we expect them to be positive. This may be explained by selection at diagnosis; individuals with certain cancers may have an increased risk of dying of other diseases in general. Then comparing with po