Download as pdf or txt
Download as pdf or txt
You are on page 1of 28

Lecture 1 - Survival Models

Lecturer: Trần Minh Hoàng

Actuarial Mathematics 1

Lecturer: Trần Minh Hoàng Lecture 1 - Survival Models Actuarial Mathematics 1 1 / 28


Table of Contents

1 The future lifetime random variable

2 Force of mortality

3 Actuarial notation

Lecturer: Trần Minh Hoàng Lecture 1 - Survival Models Actuarial Mathematics 1 2 / 28


The future lifetime random variable

Lecturer: Trần Minh Hoàng Lecture 1 - Survival Models Actuarial Mathematics 1 3 / 28


The future lifetime random variable

Let (x) denote a life aged x.


Let Tx be the future lifetime of (x). That means x + Tx is the age-at-death
for (x).
We model Tx as a continuous random variable.
Let Fx be the distribution function of Tx , so that

Fx (t) = P[Tx ≤ t].

Then Fx (t) represents the probability that (x) does not survive beyond age
x + t, and we refer to Fx as the lifetime distribution from age x.

Lecturer: Trần Minh Hoàng Lecture 1 - Survival Models Actuarial Mathematics 1 4 / 28


Survival function

Distribution of Tx can also be characterised by its survival distribution


function
Sx (t) = 1 − Fx (t) = P(Tx > t).
This is the probability that (x) survives for at least t years.

Lecturer: Trần Minh Hoàng Lecture 1 - Survival Models Actuarial Mathematics 1 5 / 28


An important assumption

We have a collection of random variables {Tx }x≥0 each has its own
distribution function.
To connect these random variables we make the following assumption:
The event {Tx ≤ t} is equivalent to {T0 ≤ x + t} given that {T0 > x}.
In other words:
P(x < T0 ≤ x + t)
P(Tx ≤ t) = P(T0 ≤ x + t|T0 > x) = .
P(T0 > x)

Lecturer: Trần Minh Hoàng Lecture 1 - Survival Models Actuarial Mathematics 1 6 / 28


An important assumption 2

Proposition
We have the following identities

S0 (x + t) = S0 (x)Sx (t),
Sx (t + u) = Sx (t)Sx+t (u).

This is a very important result. It shows that we can interpret the probability of
survival from age x to age x + t + u as the product of
the probability of survival to age x + t from age x, and
the probability, having survived from age x to age x + t, of further surviving
to age x + t + u.

Lecturer: Trần Minh Hoàng Lecture 1 - Survival Models Actuarial Mathematics 1 7 / 28


Properties of a survival function

Condition 1. Sx (0) = 1; that is, the probability that a life currently aged x
survives 0 years is 1.
Condition 2. limt→∞ Sx (t) = 0; that is, all lives eventually die.
Condition 3. The survival function must be a non-increasing function of t;

Lecturer: Trần Minh Hoàng Lecture 1 - Survival Models Actuarial Mathematics 1 8 / 28


Further assumptions

We make further assumptions on the survival distribution function to ensure Tx


has finite mean and variance.
Assumption 1. Sx (t) is differentiable for all t > 0.
Assumption 2. limt→∞ tSx (t) = 0.
Assumption 3. limt→∞ t 2 Sx (t) = 0.

Lecturer: Trần Minh Hoàng Lecture 1 - Survival Models Actuarial Mathematics 1 9 / 28


Example

Example
Suppose the distribution of the lifetime from birth random variable T0 is given by
 t 6
F0 (t) = 1 − 1 − for 0 ≤ t ≤ 120.
120
Calculate the probability that
a) a newborn life survives beyond age 30,
b) a life aged 30 dies before age 50, and
c) a life aged 40 survives beyond age 65.

Lecturer: Trần Minh Hoàng Lecture 1 - Survival Models Actuarial Mathematics 1 10 / 28


Limiting age

Note that in the previous example, survival beyond age 120 is impossible. We refer
to 120 as the limiting age of this model. In general, limiting age in a survival
model is denoted by ω.

Lecturer: Trần Minh Hoàng Lecture 1 - Survival Models Actuarial Mathematics 1 11 / 28


Force of mortality

Lecturer: Trần Minh Hoàng Lecture 1 - Survival Models Actuarial Mathematics 1 12 / 28


Force of mortality

Definition
The force of mortality at age x is denoted by µx and is defined as

P(T0 ≤ x + ∆x|T0 > x)


µx := lim + .
∆x→0 ∆x
Equivalently, µx can be defined as

P(Tx ≤ ∆x)
µx := lim + .
∆x→0 ∆x

Lecturer: Trần Minh Hoàng Lecture 1 - Survival Models Actuarial Mathematics 1 13 / 28


Interpretation of µx

The force of mortality is best understood by noting that for very small ∆x, we
have the approximation

µx ∆x ≈ P(T0 ≤ x + ∆x|T0 > x). (1)

That is for µx ∆x is approximately the probability that a life having survived x


years is to die within the next ∆x years.

Lecturer: Trần Minh Hoàng Lecture 1 - Survival Models Actuarial Mathematics 1 14 / 28


From Sx (t) to µx+t

Theorem
We can find µx and µx+t from S0 (x) and Sx (t) by

1 dS0 (x)
µx = − ,
S0 (x) dx
1 dSx (t)
µx+t =− .
Sx (t) dt

Lecturer: Trần Minh Hoàng Lecture 1 - Survival Models Actuarial Mathematics 1 15 / 28


From Sx (t) to µx+t

Corollary
Let f0 (t) and fx (t) be the probability density functions of T0 and Tx . Then

f0 (x) fx (t)
µx = , µx+t = .
S0 (x) Sx (t)

Lecturer: Trần Minh Hoàng Lecture 1 - Survival Models Actuarial Mathematics 1 16 / 28


From µx+t to Sx (t)

Theorem
We can find S0 (x) and Sx (t) from µx and µx+t by
 Z x 
S0 (x) = exp − µs ds ,
0
 Z t 
Sx (t) = exp − µx+s ds .
0

Lecturer: Trần Minh Hoàng Lecture 1 - Survival Models Actuarial Mathematics 1 17 / 28


Example: the Generalized De Moirve Law of mortality

Example
Suppose S0 (x) is given by
 x n
S0 (x) = 1 − for 0 ≤ x ≤ ω
ω

a) Find an expression for Sx (t).


b) Find an expression for µx .

Lecturer: Trần Minh Hoàng Lecture 1 - Survival Models Actuarial Mathematics 1 18 / 28


Example: Makeham’s Law of mortality

Example
Suppose µx = A + Bc x where 0 < B < 1 and c > 1.
a) Derive an expression for Sx (t).
b) Derive an expression for fx (t).
In the special case when A = 0, this model is called the Gompertz’s Law.

Lecturer: Trần Minh Hoàng Lecture 1 - Survival Models Actuarial Mathematics 1 19 / 28


Actuarial notation

Lecturer: Trần Minh Hoàng Lecture 1 - Survival Models Actuarial Mathematics 1 20 / 28


Actuarial notation

The notation used in the previous sections, Sx (t), Fx (t) and fx (t), is
standard in statistics.
Actuarial science has its own notations. It is important that we are aware of
both sets of notation but will mainly use the actuarial notation from now on.
The basic actuarial notation are (there are many more).

Actuarial symbols Definition Statistics equivalence


t px P(Tx > t) Sx (t)
t qx P(Tx ≤ t) Fx (t)
u|t qx P(u < Tx ≤ u + t) Sx (t) − Sx (t + u)

Lecturer: Trần Minh Hoàng Lecture 1 - Survival Models Actuarial Mathematics 1 21 / 28


Actuarial notation - explained

t px is the probability that (x) survives to at least age x + t,


t px is the probability that (x) dies before age x + t,
u|t qx is the probability that (x) survives u years, and then dies in the
subsequent t years, that is, between ages x + u and x + u + t.
u|t qx is called a deferred mortality probability, because it is the probability
that death occurs in some interval following a deferred period.
We may drop the subscript t if its value is 1, so that px represents the
probability that (x) survives to at least age x + 1.
Similarly, qx is the probability that (x) dies before age x + 1. In actuarial
terminology qx is called the mortality rate at age x.

Lecturer: Trần Minh Hoàng Lecture 1 - Survival Models Actuarial Mathematics 1 22 / 28


Basic Identities - t px , t qx & u|t qx

Proposition

t px + t qx = 1
u|t qx = u px − u+t px = (u px )(t qx+u ),
u+t px = (u px )(t px+u ).

Lecturer: Trần Minh Hoàng Lecture 1 - Survival Models Actuarial Mathematics 1 23 / 28


Basic Identities - t px & µx+t

Proposition

1 d t px
µx+t = − ,
t x dt
p
 Z t 
t px = exp − µx+s ds
0

Lecturer: Trần Minh Hoàng Lecture 1 - Survival Models Actuarial Mathematics 1 24 / 28


Basic Identities - µx+t , t px , t qx & fx (t)

Proposition
Z t
fx (t) = t px µx+t & t qx = s px µx+s ds.
0

Lecturer: Trần Minh Hoàng Lecture 1 - Survival Models Actuarial Mathematics 1 25 / 28


Time-line diagram for t qx

Lecturer: Trần Minh Hoàng Lecture 1 - Survival Models Actuarial Mathematics 1 26 / 28


Approximation for qx

Corollary
Z 1
qx ≈ µx+s ds ≈ µx+1/2 . (2)
0

Lecturer: Trần Minh Hoàng Lecture 1 - Survival Models Actuarial Mathematics 1 27 / 28


Example: Pareto distribution

Example
Suppose  α
λ
x p0 =
λ+x
where α, λ > 0. Calculate µx , fx (t), t px , t qx .

Lecturer: Trần Minh Hoàng Lecture 1 - Survival Models Actuarial Mathematics 1 28 / 28

You might also like