|
Genna
Overview | Genna
Uniqueness | Genna
Key features | Genna
Userbase | Genna
System Requirements
Censored
Observations
If the dependent variable
of interest represents the time to an event and
the duration of the study is limited in time, some
observations for which the event has not yet occurred
are referred to as being censored. While the concept
and analysis of censored observations owes its origins
to medical and actuarial research where the event
being modelled was generally the survival of patients
with a terminal disease or the evaluation of the
risk of insuring an individual based on certain
socio-demographic and medical factors, censored
observations may occur in a number of different
sectors.
For example, in the social sciences we may study
the "survival" of marriages, high school
drop-out rates (time to drop-out), turnover in organizations,
etc. In each case, by the end of the study period,
some subjects probably will still be married, will
not have dropped out, or will still be working at
the same company; thus, those subjects represent
censored observations.
In the telecommunications industry the event being
modelled may be churn of customers to competitors.
In this case all existing customers are viewed as
censored events and the underlying, somewhat bleak,
assumption is that existing customers are going
to churn but the event has not yet occurred.
In economics we may study the survival of new businesses,
given a set of factors that survival may be dependent
on.
In quality control, a typical goal would be the
modelling of survival of parts under stress.
back to products |
|
|
|
Coral
Summary
