ASM 09205: SURVIVAL MODELS Future lifetime Distribution
SURVIVAL MODELS
MODULE CONTENT PART I:
MODULE TOPIC
Status
Future lifetime random variables
FUTURE LIFETIME DISTRIBUTION Probabilities of death or survival
Expected future lifetime
Force of Mortality
Exponential Model
Weibull Model
Gompertz and Makeham Law of Mortality
PARAMETRIC SURVIVAL MODELS
Calculating parameter values
AND ESTIMATION OF FUTURE LIFETIME DISTRIBUTION
Calculating survival probabilities
Parametric versus non parametric models
Censoring Mechanisms
EXPOSED TO RISK Exposed to Risk
FUTURE LIFE TIME DISTRIBUTION
In this module we will discuss a model of random lifetimes where we treat the future
lifetime of an individual as a continuous random variable.
From this simple starting point we will derive many useful results that are the
building blocks of actuarial work relating to human mortality.
Although we will mostly study the lifetime model in the context of human mortality, the
theory can equally be applied to other problems, such as:
i. analyzing the lengths of time that surviving individuals hold insurance policies –
here mortality is replaced by ‘withdrawal’.
ii. analyzing the lengths of time that surviving individuals remain healthy – here
mortality is replaced by ‘sickness’.
1.1STATUS
A life insurance policy is sometimes issued and pays a benefit at a time which
depends on the survival characteristics of two or more people.
A status is an artificially constructed life form for which the notion of life and death
can be well defined.
A status can be deterministic i.e defined /numbered term or random i.e unknown
future life time. In this context, a person aged is called status and when referring
ages we may refer to age as status .
1.2 FUTURE LIFE TIME RANDOM VARIABLE
The starting point for a simple mathematical model of survival is the observation
that the future lifetime of a person (called a ‘life’ in actuarial work) is not known in
advance.
Further, we observe that lifetimes range from 0 (a new born) to in excess of 100
years (usually taken as 120) . A natural assumption therefore is that the future
lifetime of a given life is a random variable.
1.2.1Distribution function and survival function of a new-born life
1.
2.
1.2.2Distribution function and survival function of a life aged x
In insurance contexts, we will not be dealing with new-born babies, so
we need to extend the notation to deal with older individuals who has already
attained age and therefore it is important to define a random variable to be a
future lifetime after age .
The lifetime distribution then becomes conditional i.e
1.2.3 Probabilities of death and survival
We now introduce the notation used by actuaries for probabilities of death and
survival.
NOTE:
IMPORTANT EQUATIONS
Other than their meanings, you also need to know how these symbols are related to
one another.
Here are four equations that you will find very useful.
Equation 1:
This equation arises from the fact that there are only two possible outcomes: dying
within years or surviving to years from now.
Equation 2:
The meaning of this equation can be seen from the following diagram
Equation 3: Deferred probability (Probability a person aged will die between age
+ + +
It can also be stated as the probability of a person aged will survive atleast years
and then die within the next years
1.3 CURTATE FUTURE LIFETIME RANDOM VARIABLE
In many insurance applications we are interested not only in the future lifetime of an
individual, but also in what is known as the individual’s curtate future lifetime.
The curtate future lifetime random variable is defined as the integer part of future
lifetime, and is denoted by for a life aged . If we let denote the floor
function, we have
We can think of the curtate future lifetime as the number of whole years lived in the
future by an individual
For example, ⎣1⎦ = 1, ⎣4.3⎦ = 4 and ⎣10.99⎦ = 10.
We call the curtate future lifetime random variable.
1.3.1 The Probability Mass function of
Discussion Problems
1.
2.
3.
