A term usually used for unobserved individual heterogeneity. Such variation is of main concern in the medical statistics particularly in the analysis of the survival times where hazard functions can be strongly influenced by the selection effects operating in the population. There are several possible sources of this heterogeneity, the most apparent of which is that it reflects the biological differences, so that, for instance, some individuals are born with the weaker heart, or a genetic disposition for cancer. A further prospect is that the heterogeneity happens from the occured weaknesses which result from the stresses of life. Failure to take account of this kind of variation might often obscure comparisons between groups, for instance, by measures of relative risk. A simple model which attempts to permit for the variation between individuals is given as follows


 


where Z is the quantity specific to an individual, considered as the random variable over the population of individuals, and the base rate is denoted by λ(t) . What is observed in a population for which this type of model holds is not the individual hazard rate but the net result for several individuals with different values of Z.