In this chapter we show how to use several different likelihoods that are related to risk assessment, combining them with the use of a spatial field.
The first example is about survival modeling where the outcome is the time to an event. This kind of outcome is common in medical studies where time to cure or time to death are common outcomes and in industry where time to failure is a common outcome. Most studies end before all of the patients are dead, or items have failed. In this case we use a censored likelihood; the likelihood considers the accumulated probability for the individual being alive until the study ends. We will only consider a simple case of censoring, but more complicated models are possible (see ?inla.surv).