Skip to content
Tutorials Center
My learning

Topics

Lesson 15 of 43 ยท Survival Models and Life Tables

Chapter 3 review: Survival models

One-page recap of survival functions, force of mortality, life tables, fractional ages and select mortality, with practice calculators and a ten-question quiz.

Chapter 3 turned the vague idea of "how long people live" into mathematics you can compute with: a survival function, a force of mortality, a table, and rules for the spaces between whole ages. Chapter 4 multiplies these probabilities by the discount factors of chapter 2 to price insurance, so each formula here will be used again within pages.

Four cards: survival function, force of mortality, life table, life expectancy
Caption: Chapter 3 in one picture: four descriptions of the same mortality.

What you will learn

  • How the survival function, force, density and table are linked.
  • Which formula to use for each probability.
  • Which assumptions you must state when working between whole ages.

Key terms

S(x), mu(x), f(x): survival function, force of mortality, density of age at death.

t p x, t q x: survive, or die within, t years from age x.

l(x), d(x): survivors and deaths in a life table.

UDD: uniform distribution of deaths within each year of age.

Select period: years after underwriting during which mortality is lower.

The chapter in a table

Task Formula
from force to survival S(x) = exp(-integral of mu(s) ds from 0 to x)
force from survival mu(x) = -S'(x) / S(x)
survive t years t p x = S(x+t)/S(x) = l(x+t)/l(x)
die in a year q x = d(x)/l(x) = 1 - p x
survive t, die in next u (l(x+t) - l(x+t+u)) / l(x)
complete life expectancy e x = integral of t p x dt
curtate life expectancy sum over k >= 1 of k p x
UDD s q x = s q x, l(x+s) = l(x) - s d(x)
constant force s p x = (p x)^s

Lessons: survival function, life tables, select tables and fractional ages.

Try it

Check a hand calculation with a short program. The first example computes the quantities from a tiny hand-made table, the kind of exercise you will meet in exams.

Python A tiny table, checked in code

Press Run (or Ctrl + Enter) to see the result here.

Program input ยท what you'd type while the program runs

Type the answers before you press Run, one per line. Each time the program reads a line (with input()), it gets the next line from this box.

The second example verifies the integral formula for S(x) numerically, a useful habit when you work with a new force of mortality.

Python Check S(x) = exp(-integral of mu)

Press Run (or Ctrl + Enter) to see the result here.

Program input ยท what you'd type while the program runs

Type the answers before you press Run, one per line. Each time the program reads a line (with input()), it gets the next line from this box.

Common misconceptions

Dividing by the wrong l. Probabilities for a person aged x divide by l(x).

Confusing force with probability. mu(x) is a rate; q x is a probability.

Forgetting the assumption. Any fractional-age result needs UDD or constant force stated.

Check your understanding

Quick check ยท 10 questions

of 10 answered You scored / 10 ยท perfect!

  1. 1. If mu = 0.03 for all ages, what is 10 p x?

    exp(-0.03 x 10) = exp(-0.3) = 0.7408.
  2. 2. Given l(60) = 80,000 and l(70) = 64,000, 10 p 60 is

    64,000 / 80,000 = 0.8.
  3. 3. l(50) = 90,000 and l(51) = 89,400. What is q(50)?

    d(50) = 600, q(50) = 600/90,000 = 0.006667.
  4. 4. If S(x) = 1 - x/100 for 0 <= x < 100, then mu(40) equals

    mu = -S'/S = (1/100) / 0.6 = 1/60 = 0.01667.
  5. 5. With l(40) = 100, l(45) = 95, l(50) = 88 (in thousands), the probability a 40-year-old survives to 45 and dies before 50 is

    (95 - 88) / 100 = 0.07.
  6. 6. A cohort has l = 100, 90, 80, 70, 0 at ages x, x+1, x+2, x+3, x+4. The curtate expectation of life at x is

    (90 + 80 + 70 + 0) / 100 = 2.4: the sum of l at later ages divided by l(x).
  7. 7. Under UDD with q x = 0.05, the probability of dying within 0.4 years is

    s q x = 0.4 x 0.05 = 0.02 under UDD.
  8. 8. Which formula is correct for survival from the force of mortality?

    Because mu = -d ln S / dx, so S = exp(-integral of mu).
  9. 9. In a select-and-ultimate table with a 3-year select period, q [x] + 3 is

    After the select period, mortality depends only on attained age.
  10. 10. Which statement about T(x) is true?

    T(x) is continuous; K(x) = floor(T(x)) is the curtate future lifetime.

Summary

  • S(x) = exp(-integral of mu); t p x = S(x+t)/S(x); q x = 1 - p x.
  • A life table gives all probabilities as ratios of l values.
  • Life expectancy: complete = integral of t p x, curtate = sum of k p x.
  • Fractions of a year need UDD or constant force; select tables reflect underwriting.

Next: Chapter 4 overview: Life insurance benefits