ABSTRACT
The non-parametric methods described in Chapter 2 can be useful in the analysis of a single sample of survival data, or in the comparison of two or more groups of survival times. However, in most medical studies that give rise to survival data, supplementary information will also be recorded on each individual. A typical example would be a clinical trial to compare the survival times of patients who receive one or other of two treatments. In such a study, demographic variables such as the age and sex of the patient, the values of physiological variables such as serum haemoglobin level and heart rate, and factors that are associated with the lifestyle of the patient, such as smoking history and dietary habits, may all have an impact on the time that the patient survives. Accordingly, the values of these variables, which are referred to as explanatory variables, would be recorded at the outset of the study. The resulting data set would then be more complex than those considered in Chapter 2, and the methods described in that chapter would generally be unsuitable.
