ABSTRACT
Survival models are interpreted as the possibility that, at a given moment, the event of interest occurs given that it has not occurred yet. This possibility is named “hazard rate”. The coefficients of the survival models are often interpreted as “hazard rates”, which is the quotient of the probability of the event happening and the survival function. There are two types of survival models, the so-called parametric models and the semi-parametric models. The former corresponds to those that make assumptions about the characteristics of the population to which the sample is representative of. The most common survival model for this category is Weibull's model. Semi-parametric models do not make any type of assumptions about the base function, as this is estimated from the data. The most famous example of this specification is Cox's model. In addition to Gantt charts, it is common that people who work with survival models show graphs of survival curves comparing two groups of interest.
