Chapter 0 and Chapter 1 Flashcards

1
Q

Define survival analysis

A

Study of the distirbution of lifetimes or time from some initating event to a terminal event - can be applied to study of distribution of times from changing one state to another

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

Where are survival models used in actuarial

A

Life assurance
Annuties
Critical illness cover
Pension funds

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

Why is the difference in actuarial vs statistical notation so important?

A

In survival models actuaries are interested in financial significance vs statisticians statistical significance - something can be acceptable under actuarial criteria but not statistically so its key we understand the purpose of the model

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

Explain meaning of variable T

A

The future lifetime of a new born person is a random variable T which is continuously distirbuted over interval 0 to omega where omega is between 0 and infinity

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

Explain limiting age

A

Denoted omega - usually 100 - 200 years old and is the oldest or max age to live to.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

Explain meaning of variable Tx

A

The future life of a person aged x is a random variable T describing future lifetime after age x of person who survives until age x. T0= T by convention

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

What is the only distirbution function we need to find all Fx(t)

A

F0(t)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

What does tqx mean?

A

Rate of mortality : Probability of life age x dying before age x+t

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

What does tpx mean?

A

Probability of life age x surviving next t years

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

What does qx mean

A

Probability that a life alive at age x dies before age x+1

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

What is the probability of surviving for time s+t after age x? in terms of multiplying two probabilities

A

Multiplying probability of surviving for time s after age x and probability of then surviving for a further time t at age x+s

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

Desribe the force fo mortality in words

A

The instantaneous measure of mortality at age x - as in you’re alive at x but die instantly. This measure can be over 1 briefly for short periods of time

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

Define the complete expectation of life

A

The complete expectation of life at age x is defined as E(Tx) which is the expected future lifetime after age x. Often it is used to measure countries’ standard of living however this average can be skewed so the median would be a better measure.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

Define the curtate expectation of life

A

Curtate expectation of life is the expectation of Kx random variable where Kx is age last birthday. IE E[Tx] - rounded down

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

How can the Curtate expectation of life be an non integer value

A

It is average. although Kx has to be an integer the average of a collection of integers can be a decimal

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

IF i have the force of mortality what can I work out about survival

A

Survival rates through lemma 5 or survival function

17
Q

Why is the exponential model with constant hazard not realistic for humans

A

Mortality rates are independent of age in this model - not true in reality

18
Q

How does the Weibull model relate to the exponential

A

If Beta = 1 then the hazard is constant and weibull model is the same as the exponential. However unlike exponential different values of beta alter the hazard to be monotonically increasing or decreasing with time

19
Q

Discuss Gompertz law

A

Says with age force of mortality goes up and keeps going up, its entirely driven by age. This is a reasonable assumption for middle to older age groups.

20
Q

Discuss makehams law

A

Includes a constant figure that can be interpreted as an allowance for accidental deaths not depending on age. - good approximation

21
Q

What are life tables

A

These are routinely published by the profession - they summarise the mortality experience of a certain group of lives. The tables are usually refreshed with each census and there are lots of different tables with different uses

22
Q

What does x denote int he tables

A

The exact age of the person on their last birthday

23
Q

What does lx denote int he tables

A

Number of people surviving to exact age x out of original 100k aged 0

24
Q

What does dx denote in the tables

A

number of deaths in year age x to x+1 out of the lx persons who enter that year

25
Q

What does Lx denote in the tables

A

Population to be expected according to the Life Table aged between x and x+1 years, assuming deaths occur evenly over the year

26
Q

What does Tx denote in the tablest

A

Expected number of person years to be lived byt eh survivors at age x

27
Q

What does e^ox denote in the tables

A

Life expectancy at age x for each person surviving, or total future lifetime in years which an average is passed through by people aged exactly x - period life expectancy

28
Q

Explain life expectancy

A

Average number of years a person would live if current mortality rates applied in the future.

29
Q

Explain the difference between period life expectancy and cohort life expectancy

A

Period life expectancy assumes mortality rates remain constant into the future, while cohort life expectancy uses projected changes in future mortality rates