www.vagupu.

com

Chapter 1 Continuous Probability Distributions

1 CONTINUOUS
PROBABILITY
DISTRIBUTIONS

Objectives
After studying this chapter you should
• know the definition of E(X) and E(g(X)) for both discrete and
continuous distributions;
• know the definition of mean and variance in terms of
expectations;
• be able to do calculations involving linear combinations of
independent normal random variables;
• be able to calculate probabilities using the exponential
distribution.

1.0 Introduction
In the text Statistics, continuous probability distributions were
introduced in Chapter 7. Probability was represented by the
area under a curve, known as the probability density function.
A probability density function of a random variable, X, must be
non-negative for all values of X and the total area under the
curve must be 1.

If the probability density function is denoted f ( x ) then the

probability that an observed value of X lies between a and b is
given by
b
P( a < x < b ) = ∫ f ( x )dx
a
a b x

Also note that

∞
∫−∞
f ( x )dx = 1.

The most important continuous distribution and indeed the most

important distribution of any kind in statistics is the normal
distribution. This was introduced in Statistics, Chapter 8.
A familiarity with the normal distribution is essential for an
understanding of what follows.

1
 www.vagupu.com
Chapter 1 Continuous Probability Distributions

This chapter starts by drawing together definitions and laws of

expectations and variances for discrete and continuous
distributions. Many of the ideas have already been met in
Statistics. Some well known results are proved. Students may
prefer to omit the proofs, particularly at first reading. It is
possible to leave sections of this chapter, which deal with some
of the more mathematical ideas in Statistics, until later in the
course if preferred.

1.1 Expectation
For a discrete probability distribution

E ( X ) = ∑ xP( X = x )
x

If g( x ) is a function of x then

E ( g( x )) = ∑ g( x )P( X = x )
x

If, for example, the random variable X can take the values 1, 2 or
3 with probabilities 0.2, 0.3 and 0.5 respectively, then

E ( X ) = 0.2 × 1 + 0.3 × 2 + 0.5 × 3 = 2.3

E ( X ) = 0.2 × 1 + 0.3 × 2 + 0.5 × 3 = 1. 49 .

For a continuous distribution

∞
E( X ) = ∫
−∞
x f ( x )dx

∞
and E ( g ( X )) = ∫ g( x ) f ( x )dx
−∞

Example
The random variable X has probability density function

x (2 − x ) for 0 < x < 2 Find E ( X ) and E   .

3 1
4  X

Solution
2
∫
3 2
E( X ) = x ( 2 − x )dx
0 4

∫ (2x )
3 2
= 2
− x 3 dx
4 0

2
 www.vagupu.com
Chapter 1 Continuous Probability Distributions

2
3 2 x4
=  x3 − 
4 3 4 0

3  16
= − 4
4 3 

=1

E  =
2
∫
1 1 3
and . x (2 − x )dx
 X 0 x 4

2
∫
3
= (2 − x )dx
0 4

2
3 x2
= 2x − 
4 2 0

= 1.5

1.2 Mean and variance

For any random variable whether continuous or discrete, the
mean is defined to be E(X) and this is usually denoted by µ .
The variance is defined to be

E ( X − E ( X ))
2

and this is usually denoted σ 2

(
σ2 = E [X − µ]
2
)
An equivalent formula is

( )
σ2 = E X2 − µ 2 (see section 1.4)

Why is the second formula for σ 2 preferred for calculations?

The next example is to revise some of the ideas introduced in

the text Statistics, Chapter 7.

Example
A charity group raises funds by collecting waste paper. A skip-
full will contain an amount, X, of other materials such as plastic
bags and rubber bands. X may be regarded as a random variable

3
 www.vagupu.com

Chapter 1 Continuous Probability Distributions

with probability density function

f ( x ) = k ( x − 1)( 4 − x ) 1< x < 4

=0 otherwise.

(All numerical values in this question are in units of 100 kg.)

2
(a) Show that k = .
9
(b) Find the mean and standard deviation of X.
(c) Find the probability that X exceeds 3.5.

A skip-full may normally be sold for £250 but if X exceeds 3.5

only £125 will be paid. Find the expected value of a skip-full.
Alternatively the paper may be sorted before being placed in the
skip. This will ensure a very low value of X and a skip-full may
then be sold for £310. However the effort put into sorting means
that 25per cent fewer skips-full will be sold. Advise the charity
whether or not to sort the paper. (AEB)

Solution
4
(a) ∫ k ( x − 1)(4 − x )dx = 1
1

∫ (− x )
4
k 2
+ 5x − 4 dx = 1
1

4
 − x 3 5x 2 
k + − 4x  = 1
 3 2 1

8 11 
k − −   = 1
3  6  
4.5k = 1
9 2
k =1 ⇒ k =
2 9
4
∫
2
(b) Mean, E( X ) = x ( x − 1)( 4 − x )dx
1 9
2
− x 3 + 5x 2 − 4x  dx
4
=
9 ∫ 1 
4
2  − x 4 5x 3 4x 2 
=  + −
9 4 3 2 1

2  32  7  
=  − − 
9  3  12  
= 2.5

4
 www.vagupu.com
Chapter 1 Continuous Probability Distributions

( ) ∫
4 2 2
E X2 = x ( x − 1)( 4 − x )dx
1 9

∫ (− x )
2 4
= 4
+ 5x 3 − 4x 2 dx
9 1

4
2  x 5 5x 4 4x 3 
= − + −
9  5 4 3 1

2  448  17  
=  − − 
9  15  60  

= 6. 7

Hence

( )
variance = E X 2 − µ 2 = 6. 7 − 2.52 = 0. 45
and

standard deviation = variance = 0.671 .

(c) Probability that X exceeds 3.5 is given by

4
∫
2
P( X > 3.5) = ( x − 1)( 4 − x )dx
3.5 9
4
2  x 3 5x 2 
=  − + − 4x 
9 3 2  3.5

2  8 7
= −
9  3 3
= 0.0741

(d) Expected value of a skip-full

= £125 × 0.0741 + £250 × (1 − 0.0741)
= £241
For each skip-full of unsorted material only 0.75 skips-full
of paper would be obtained after sorting. This would be
worth
0. 75 × £310 = £232.5

Thus the charity will make more money in the long run if
the paper is not sorted.

5
 www.vagupu.com
Chapter 1 Continuous Probability Distributions

Exercise 1A
1. A temporary roundabout is installed at a crossroads. 3. A technique for measuring the density of a
The time, X minutes, which vehicles have to wait silicon compound is a random variable, X, with
before entering the crossroads has probability probability density function
density function
f (x) = k −0.04 < x < 0.04
f ( x ) = 0.8 − 0.32 x 0 < x < 2.5 =0 otherwise
=0 otherwise (a) Find the value of k.
Find the mean and standard deviation of X.
(AEB) (b) Find the mean and standard deviation of X.
2. The random variable, Y, has probability density (c) Find the probability that the error is between
function −0.03 and 0.01.
f ( y ) = k (8 − 2 y ) 0<x<4 (d) Find the probability that the magnitude of the
=0 otherwise error is greater than 0.035.
(a) Verify that k = 0.0625 and that the median
is 1.172.
(b) Find the mean and standard deviation of Y.
(c) What is the probability that the mean of a
random sample of size 50 from this
distribution will lie between the mean and
the median? (AEB)

1.3 Laws of expectation

The following laws apply to expectations of both discrete and
continuous random variables.

(i) E ( a + bX ) = a + bE ( X )

where a and b are constants and X is a random variable.

(ii) E ( X + Y ) = E ( X ) + E (Y )

where X and Y are any random variables.

Laws (i) and (ii) may be extended to a more
general form

(iii) E ( a1 X1 + a2 X2 + K + an Xn )
= a1 E ( X1 ) + a2 E ( X2 ) + K + an E ( Xn )

when X1 , X2 K, Xn are random variables and

a1 , a2 K, an are constants.

(iv) E ( XY ) = E ( X ) E (Y )

when X and Y are independent random variables .

Note that (iv) is only true if X and Y are independent.

6
 www.vagupu.com

Chapter 1 Continuous Probability Distributions

1.4 Some results for variances

The laws of expectation together with the definition of variance
can be used to derive some interesting and useful results. In the
following V(X) is used to denote the variance of X and E ( X ) = µ
and V ( X ) = σ 2 .

Note: Brief proofs are included but will not be tested in the AEB
examination and may be omitted.

(i) (
V(X) = E [X − µ ]
2
)
( )
= E X 2 − 2µE ( X ) + µ 2

= E( X ) − 2µ2 2
+ µ2

= E( X ) − µ 2 2

This result has been used in Statistics and in

Section 1.2 above.

(ii) V ( a + bX ) = b 2 V ( X )

where a and b are constants.

Since
E ( a + bX ) = a + bE ( X )
= a + bµ ,
then

(
V ( a + bX ) = E [ a + bX − a − bµ ]
2
)
= E(b [ X − µ ] )
2 2

= b E ([ X − µ ] )
2 2

= b 2V ( X )

(iii) If X and Y are independent random variables

V ( X ± Y ) = V ( X ) + V (Y )

Note: (a) this is only true if X and Y are independent;

(b) the left hand side has a sign but the variances are
always added. Clearly it would not make sense to
take one variance from another as this could lead to a
negative result. A negative variance is impossible.

7
 www.vagupu.com

Chapter 1 Continuous Probability Distributions

The proof follows from

E ( X + Y ) = E ( X ) + E (Y )
= µ x + µ y , say

[ {
V ( X ± Y ) = E X ± Y − µ x ± µ y }] 
2

[ {
= E {X − µ x } ± Y − µ y }] 
2

( ) [
= E [ X − µ x ] + E  X − µ y
2
]
2
( [ ])
± 2E [ X − µ x ] Y − µ y 

= V ( X ) + V (Y ) ± 2E ( X − µ x ) E (Y − µ x )

since X and Y are independent.

Now
E( X − µ x ) = E( X ) − µ x
= µx − µx
=0
Hence
V ( X ± Y ) = V ( X ) + V (Y ) .

(iv) The results above can be extended to

V ( a1 X1 + a2 X2 + K + an Xn )

= a12 V ( X1 ) + a2 2 V ( X2 ) + K + an 2 V ( Xn )

where a1 , a2 , K an are constants and X1 , X2 , K Xn are

independent random variables.

1.5 Distribution of the sample

mean
The results above can be used to derive the mean and variance
of a sample mean.

(a) If X is the mean of a random sample of n observations of

the random variable X, then

 [ X + X2 + K + Xn ] 
E( X ) = E 1 
 n 

8
 www.vagupu.com

Chapter 1 Continuous Probability Distributions

where Xi denotes the ith observation in the sample, giving

E( X ) =
1
n
(
E[ X1 ] + E[ X2 ] + K + E[ Xn ] )
=
1
(µ + µ + K + µ )
n

=
1
(nµ )
n
=µ

(b) V ( X ) = E ( X − µ ) since the mean of X has just been shown

2

to be µ .

V ( X ) = V  ( X1 + X2 +... + Xn )
1
n 

=
1
n2
(
V ( X1 ) + V ( X2 ) + K + V ( Xn ) )
=
1
n2
(
σ 2 + σ 2 + K+σ 2 )
=
1
n2
(
nσ 2 )
σ2
=
n
You have now shown that the sample mean X is distributed
σ2
with mean µ and variance , a result which has been
n
used frequently in Statistics.

1.6 Unbiased estimate of

variance
The results above can be used to find an unbiased estimate of
the variance.

First consider

([
∑ ( Xi − µ ) = ∑ Xi − X + X − µ ] [ ])
2 2

= ∑ ( Xi − X ) + ∑ ( X − µ ) + 2 ∑ ( Xi − X )( X − µ )
2 2

9
 www.vagupu.com

Chapter 1 Continuous Probability Distributions

Now ∑ ( Xi − X )( X − µ ) = ( X − µ )∑ ( Xi − X )

(
= ( X − µ ) [ ∑ Xi ] − nX )
= ( X − µ )( ∑ Xi − ∑ Xi )

=0

Hence

∑ ( Xi − µ ) = ∑ ( Xi − X ) + ∑ ( X − µ )
2 2 2

= ∑ ( Xi − X ) + n ( X − µ ) .
2 2

Taking expectations of both sides gives

(
∑ E [ Xi − µ ]
2
) = E(∑ [ X − X ] ) + nE([ X − µ ] )
i
2 2

( [ ] ) + nnσ
2
2
⇒ nσ 2 = E ∑ Xi − X .

Hence

( [
E ∑ Xi − X ] ) = (n − 1)σ 2
2

or
( [
E ∑ Xi − X ])
2

= σ2 .
( n − 1)

The expression on the left hand side was met in Statistics and is
( )
denoted σ̂ 2 . You have just shown that E σ̂ 2 = σ 2 . σ̂ 2 is said
to be an unbiased estimator of σ and this is why it is generally
2

used to estimate σ 2 . Note however that σ̂ is not an unbiased

estimator of σ .

Why is σ̂ not an unbiased estimator of σ ?

10
 www.vagupu.com

Chapter 1 Continuous Probability Distributions

1.7 Distribution of a linear

combination of independent
normal random variables
The results in Sections1.3 and 1.4 cover the mean and variance
of linear combinations of independent random variables. If the
variables are all normally distributed then a linear combination
of them will also be normally distributed. The proof of this is
beyond the scope of this book.

To take a simple example, in the mass production of jars of jam

the weight of jam put in each jar is a normally distributed
random variable, X, with mean 456 g and standard deviation 4
g. The weight of the jar (including the lid) is an independent
normally distributed random variable, Y, with mean 35 g and
standard deviation 3 g.

The total weight of the jam plus the jar is X + Y .

The mean of X + Y = E ( X + Y ) = E ( X ) + E (Y ) = 456 + 35 = 491 g

The variance of X + Y = V ( X ) + V (Y ) = 4 2 + 32 = 25 .

The standard deviation will be 25 = 5 g.

Since the two variables are independently normally distributed

the distribution of the total weight will also be normal.

A child opens a new jar of jam and takes a spoonful out. The
weight of jam in the spoon is a random variable, Z, with mean
22 g standard deviation 2 g. Z is independent of X.
The weight of jam remaining in the jar is X − Z .

The mean is E ( X ) − E ( Z ) = 456 − 22 = 434 g.

The variance of X − Z is V ( X ) + V ( Z ) = 4 2 + 2 2 = 20 .

The standard deviation is 20 = 4. 47 g.

Since X and Z are independently normally distributed the
distribution of X − Z is also normal.

Example
A machine produces rubber balls whose diameters are normally
distributed with mean 5.50 cm and standard deviation 0.08 cm.

(a) What proportion of balls will have diameters

(i) less than 5.60 cm,
(ii) between 5.34 and 5.44 cm?

11
 www.vagupu.com

Chapter 1 Continuous Probability Distributions

(b) The balls are packed in cylindrical tubes whose internal

diameters are normally distributed with mean 5.70 cm and
standard deviation 0.12 cm. If a ball, selected at random, is
placed in a tube, selected at random, what is the distribution of
the clearance? (The clearance is the internal diameter of the
tube minus the diameter of the ball.) What is the probability
that the clearance is between 0.05 cm and 0.25 cm?
(AEB)

Solution
(5.60 − 5.50 )
(a) (i) z= = 1.25 . 0.89435
0.08
Using Normal tables, the probability of being less than
5.60 cm is 0.894.
1.25
(5.34 − 5.50 )
(ii) z1 = = −2.0
0.08
(5. 44 − 5.50 )
z2 = = −0. 75 .
0.08
Hence the probability of being between 5.34 cm and 1 – 0.77337

5.44 cm is given by 1 – 0.97725

0.97725 − 0. 77337 = 0.204 . –2 –0.75

(b) If X is the diameter of the ball and Y is the diameter of the

tube the clearance is Y − X . This will be normally distributed
with
mean = 5. 70 − 5.50 = 0.20

variance = 0.082 + 0.12 2 = 0.0208

0.6357
standard deviation = 0.1442.

To calculate the probability that the clearance is between

0.05 cm and 0.25 cm you first find 1 – 0.8508

(0.05 − 0.20 )
–1.04 0.347
z1 = = −1.040
0.1442

(0.25 − 0.20 )
z2 = = − 0.347
0.1442

so that probability of the clearance between 0.05 cm and

0.25 cm is given by

0.6357 − (1 − 0.8508) = 0. 4865 .

Note: Interpolation has been used in reading the normal tables, but
the effect on the final answer is small.

12
 www.vagupu.com

Chapter 1 Continuous Probability Distributions

Example
A baker makes digestive biscuits whose masses are normally
distributed with mean 24.0 g and standard deviation 1.9 g. The
biscuits are packed by hand into packets of 25.

(i) Assuming the biscuits included in each packet are a random

sample from the population, what is the distribution of the
total mass of biscuit in a packet and what is the probability
that it lies between 598 g and 606 g?
(ii) Ten packets of biscuits are placed in a box. What is the
probability that the total mass of biscuit in the box lies
between 6010 g and 6060 g?
(iii) A new packer was including 26 biscuits in each packet.
What is the probability that a packet selected at random from
those containing 25 biscuits would contain a greater mass of
biscuits than a packet selected at random from those
containing 26 biscuits? (AEB)

Solution
(i) Since the sample is random the masses of the 25 biscuits
included in the packet will be independent of each other.
The distribution of the total mass will therefore be normal.

The mean mass of the biscuits in the packet will be

24 + 24 + K + 24 = 25 × 24 = 600 g.

The variance of the total mass will be

1.9 2 + 1.9 2 + K + 1.9 2 = 25 × 1.9 2 = 90.25,
giving a standard deviation = 9.5 g.

(
The distribution is N 600, 9.52 . )
To find the probability the total mass lies between 598 g and
606 g, note that

(598 − 600 ) 0.7363

z1 = = −0.211
9.5
1 – 0.5836

(606 − 600 )
z2 = = 0.632 .
9.5 –0.211 0.632
Probability between 598 g and 606 g

= 0. 7363 − (1 − 0.5836 ) = 0.320.

(ii) Ten packets of biscuits are placed in a box. Provided their
masses are independent the total mass will be normally
distributed with mean 10 × 600 = 6000 g.
The variance will be 10 × 90.25 = 902.5 and the standard
deviation 902.5 = 30.04 g. To calculate the probability
between 6010 g and 6060 g,

13
 www.vagupu.com
Chapter 1 Continuous Probability Distributions

(6010 − 6000 )
z1 = = 0.333
30.04 0.9771
0.6304
(6060 − 6000 )
z2 = = 1.997 .
30.04
0.333 1.997
Probability of being between 6010 g and 6060 g
= 0.9771 − 0.6304 = 0.347 .

(iii) If X is the mass of biscuit in a packet containing 25 biscuits

then we already know that X is N(600, 90.25). If Y is the mass
of biscuit in a packet containing 26 biscuits then Y is normally
distributed with mean 26 × 24 = 624 g and variance
26 × 1.9 2 = 93.86 .
The difference Y – X is normally distributed with

mean = 624 − 600 = 24 g,

variance = 90.25 + 93.86 = 184.11 ,

0.9615
standard deviation = 184.11 = 13.57 .

A packet containing 25 biscuits will weigh more than a packet

containing 26 biscuits if Y – X is negative, i.e. Y − X < 0. –1.769

Thus
( 0 − 24)
z= = −1. 769 .
13.57
This gives the probability of this occurring as

1 − 0.9615 = 0.0385 .

Exercise 1B
1. A dispenser discharges an amount of soft drink 2. The weights of pieces of home made fudge are
which is normally distributed with mean 475 ml normally distributed with mean 34 g and
and standard deviation 20 ml. standard deviation 5 g.
(a) What is the distribution of the total amount (a) A bag contains 15 pieces of fudge chosen at
in two independent drinks? random. What is the distribution of the total
(b) If the capacity of the cups into which the weight of fudge in the bag? What is the
drink is dispensed is normally distributed probability that the total weight is between
with mean 500 ml and standard deviation 490 g and 540 g?
30 ml, what is the distribution of the (b) What is the probability that the total weight
difference between the capacity of a cup of fudge in a bag containing 15 pieces
and the amount dispensed? exceeds that in another bag containing
Assume the two are independent. 16 pieces? (AEB)

14
 www.vagupu.com
Chapter 1 Continuous Probability Distributions

3. Audrey is a regular customer of Toto’s taxis. (b) Find the mean of Y, assuming a normal
When she rings from home the time, X, a taxi distribution.
takes to arrive is normally distributed with mean
19 minutes and standard deviation 3 minutes. (c) What is the distribution of T where T = X − Y ?
(X and Y may be assumed independent.) If
(a) (i) What is the probability of her having to both firms were rung at the same time, what
wait less than 15 minutes for a taxi? is the probability that Toto would arrive first?
(ii) What waiting time will be exceeded with (d) In order to catch a train Audrey needs a taxi
probability 0.1? within 10 minutes. Which firm would you
Audrey decides to try Blue Star taxis. The advise her to ring? Explain your answer.
standard deviation of her waiting time, Y, is (AEB)
7 minutes and the probability of Y exceeding
8 minutes is 0.97725.

1.8 The exponential distribution

If cars passing a point on a motorway follow a Poisson
distribution, the interval of time between successive cars
passing follows an exponential distribution.

The exponential distribution has probability density function

me − mx x > 0 , where m is constant

f (x) = 
0 otherwise

The mean of this exponential distribution is

∞
 e − mx 
[ ] ∫
∞ ∞ ∞
∫
1
mxe − mx dx = − xe − mx − −e − mx dx =  −  =m
0 0 0  m 0

( ) ∫ [ ] −∫
∞ ∞ ∞ 2
E X2 = mx 2 e − mx dx = − x 2 e − mx −2xe − mx dx =
0 0 0 m2

since we have already shown that

∞
∫
1
mxe − mx dx = .
0 m

Thus
2 1 1
V(X) = − = .
m2 m2 m2

1
Hence the standard deviation is .
m

Note: the AEB syllabus requires polynomial integration only

and these proofs will not be tested.

15
 www.vagupu.com

Chapter 1 Continuous Probability Distributions

1.9 Cumulative distribution

function
The probability that an observed value from the negative
exponential distribution is less than x is given by

[ ]
x
P( X < x ) = ∫
x
me − mx dx = −e − mx = 1 − e − mx
0 0

This expression is known as the cumulative distribution function.

It is denoted F(x).

For this distribution

F ( x ) = 1 − e − mx

The cumulative distribution function may be used as an alternative

to integration for evaluating probabilities. If a and b are two
constants and a > b , the probability that X takes a value between
a and b is

P( X < a ) − P( X < b ) = F ( a ) − F ( b )

For example, the interval, X seconds, between cars passing a point

on a motorway follows an exponential distribution with probability
density function

2e −2 x 0<x
f (x) = 
0 otherwise

(This distribution may alternatively be described as an exponential

distribution with parameter 2.)

The probability that the next interval is between 1 and 2 seconds is

( )
F ( 2 ) − F (1) = 1 − e −4 − 1 − e −2 = 0.9817 − 0.8647 = 0.117 .

You may check this result by evaluating

2
∫
1
2e −2 x dx .

The probability that the next interval is longer than 3 seconds is

(
F ( ∞ ) − F (3) = 1 − 1 − e −6 = 0.0025 . )
The probability that the next interval is less than 1.5 seconds is

( ) (
F (1.5) − F ( 0 ) = 1 − e −3 − 1 − e 0 = 0.9502 . )

16
 www.vagupu.com

Chapter 1 Continuous Probability Distributions

1
This exponential distribution will have a mean of = 0.5 seconds.
2
Therefore the average number of cars passing the point per second
1
is = 2.
0.5

In general the intervals between successive events from a Poisson

distribution with mean m are distributed according to the
exponential distribution with parameter m.

1.10 Conditional probability and

the exponential distribution
Perhaps the most interesting feature of the exponential distribution
is that it is 'memory less'. Using the cumulative distribution for
the exponential distribution with parameter m, the probability that
the next interval will last longer than a seconds is

( )
F ( ∞ ) − F ( a ) = 1 − 1 − e − ma = e − ma .

The probability that the interval lasts longer than a + b seconds is

similarly e − m ( a + b ) .

The probability that an interval lasts longer than a + b , given that

it has already lasted a seconds, may be written

P( X > a + b X > a ) .

Using the laws of probability

[ ]
P ( X > a + b ) ∩ ( X > a ) = P( X > a ) × P( X > a + b X > a ) .

If an interval lasts longer than a + b seconds it must have lasted

longer than a seconds. Therefore

[ ]
P ( X > a + b ) ∩ ( X > a ) = P( X > a + b ) = e − m ( a + b )

Therefore e − m ( a + b ) = e − ma P( X > a + b X > a ) .

i.e. P( X > a + b X > a ) = e − mb = P( X > b ) .

That is, the probability of the next event occuring within

b seconds is the same whether an event has just occurred or
whether the last event occurred a seconds ago.

Why is a distribution with this probability described as 'memory

less'?

17
 www.vagupu.com
Chapter 1 Continuous Probability Distributions

In the earlier example the probability that the interval between

cars passing a point on a motorway would be longer than 3
seconds was calculated. The calculation would have been
identical if the probability required had been that no car will
pass in the next 3 seconds.

Exercise 1C
1. The interval, T minutes, between successive 3. The lives of electric light bulbs, T hours, follows
telephone calls to a school office follows an an exponential distribution with probability
exponential distribution with parameter 0.2. distribution function
Find the probability that the interval between the
next two telephone calls will be  0.004e −0.004t 0<t
f (t ) = 
(a) between 3 and 6 minutes, 0 otherwise
(b) between 2 and 7 muinutes, What is the probability that a bulb will last
(c) longer than 8 minutes, (a) between 200 and 300 hours,
(d) less than 10 minutes. (b) between 250 and 350 hours,
2. A factory worker is employed to watch a monitor (c) more than 400 hours
and give a warning when the monitor signals that
(d) less than 100 hours?
action is needed to adjust the process. The
interval, X hours, between successive signals What is the mean and standard deviation of T?
follows an exponential distribution with What is the mean and standard deviation of the
parameter 0.08. mean life of 225 bulbs?
What is the probability that the interval between Find, approximately, the probability that the
the next two signals is mean life of 225 bulbs will be less than
(a) between 10 and 20 hours, 240 hours.
(b) between 6 and 16 hours
(c) longer than 50 hours,
(d) less than 2 hours?
The factory worker decides to read a newspaper
for a few minutes instead of watching the
monitor. How long can she read for if the
probability of missing a signal is to be less
than 0.01?

1.11 Miscellaneous Exercises

1. A clothing factory uses rolls of cloth for making (b) Write down the mean and variance of X , the
suits. The length, in cm, of cloth wasted mean of a sample of size n from the
(because it is too short to use) at the end of each distribution.
roll may be regarded as a random variable, X, (c) The median, Y, of a random sample of size 3
with probability density function from the distribution has a probability
density function
1 0<x<a

f (x) =  a  6 y 6 y2
 0  − 3 0<y<a
otherwise g( y ) =  a 2 a
0
(a) Derive the mean, µ ,and variance, σ 2 , in  otherwise
terms of a. Find the mean and variance of Y. (AEB)

18
 www.vagupu.com

Chapter 1 Continuous Probability Distributions

2. The Sunset Times has an average of 1 typing (a) Find the probability that the total mass of the
error per 500 words. The distribution of X, the sealed tin and its beans
number of words between successive typing
(i) exceeds 550 g,
errors, has probability density function
(ii) lies between 466 g and 575 g.
 0.002e −0.002x 0<x
f (x) = 
(b) Calculate an interval within which
0 otherwise
approximately 90% of the masses of the filled
Find the probability that the number of words tins will lie.
between successive errors is
The tins are packed in boxes of 24, the mass
(a) between 400 and 600, of the box being normally distributed with
(b) between 200 and 800, mean 500 g and standard deviation 30 g.

(c) greater than 1000, (c) Find the probability that a full box weighs
less than 12.75 kg.
(d) less than 100.
5. Lin Ying belongs to an athletic club. In 800 m
What is the probability that a reader starting at a races her times are normally distributed with
random point in the paper will read more than mean 128 seconds and standard deviation
750 words before reaching an error? 4 seconds.
The editor, Mr B. Pad, always reads the first (a) What is the probability of her time in an 800
paragraph before the paper is published. How being between 120 and 130 seconds?
many words can the first paragraph contain if
the probability of it containing an error is to be (b) What time will she beat in 70% of her races?
not more than 0.05? Julie belongs to the same club. In 800 m races
3. The thickness of a certain grade of hardboard 85% of her times are less than 140 seconds and
stocked by a DIY shop is normally distributed 70% are less than 135 seconds. Her times are
with mean 7.3 mm and standard deviation normally distributed.
0.5 mm.
(c) Find the mean and standard deviation of
(a) What proportion of sheets will be between Julie's times, each correct to two significant
7 mm and 8 mm thick? figures.
(b) Sheets of the same grade bought from a (d) What is the probability that in an 800 m race
second shop contain 9.1% over 8 mm thick Lin Ying will beat Julie?
and 2.3% less than 7 mm thick. Assuming
that the thickness is normally distributed, (e) The club has to choose one of these two
find its mean and standard deviation correct athletes to enter a major competition. In
to the nearest tenth of a mm. order to qualify for the final rounds it is
necessary to achieve a time of 114 seconds or
(c) What is the distribution of Y – X where X and less in the heats. Which athlete should be
Y are the thickness of pieces of hardboard chosen and why?
selected at random from the first and second
shops respectively? Find the probability that 6. The contents of bags of oats are normally
X exceeds Y. distributed with mean 3.05 kg, standard deviation
0.08 kg.
(d) It is possible to buy batches of hardboard
from the first shop with any required mean (a) What proportion of bags contain less than
and with the standard deviation remaining 3.11 kg?
0.5 mm. What value of the mean should be (b) What proportion of bags contain between 3.00
chosen and 3.15 kg?
(i) to minimise the proportion of sheets (c) What weight is exceeded by the contents of
outside the range 7 mm to 8 mm (no 99.9% of the bags?
proof required),
(ii) so that 0.1% of sheets are less than (d) If 6 bags are selected at random, what is the
7 mm thick? probability that the mean weight of the
contents will be between 3.00 and 3.15 kg?
4. A certain brand of beans is sold in tins, the tins
being filled and sealed by a machine. The mass The weight of the bags when empty is normally
of beans in each tin is normally distributed with distributed with mean 0.12 kg, standard deviation
mean 425 g and standard deviation 25 g and the 0.02 kg. Full bags are packed into boxes each of
mass of the tin is normally distributed with mean which holds 6 bags.
90 g and standard deviation 10 g.

19
 www.vagupu.com
Chapter 1 Continuous Probability Distributions

(e) What is the distribution of the weight in a

box, i.e. 6 bags together with their contents?
Assume that the weight of all bags and
contents in a box are independent of each
other.
(f) Within what limits will the weight in a box
lie with probability 0.9?
7. Fertilizer is packed, by a machine, into bags of
nominal mass 12 kg. The random mass of each
bag may be regarded as a normally distributed
random variable with mean 12.05 kg, standard
deviation 0.20 kg.
(a) What is the probability that the mass of a bag
exceeds 12 kg?
(b) What mass is exceeded by exactly 95% of the
bags?
A farmer buys 20 bags at a time.
(c) What is the probability that their mean mass
will exceed 12 kg?
(d) What is the distribution of the total mass of
the 20 bags and what is the probability that it
lies between 239.5 kg and 240.5 kg?
The mass of the bags packed by a second
machine may be regarded as a normally
distributed random variable with mean 12.05 kg,
standard deviation 0.05 kg. If the farmer's 20
bags are made up of n from the first machine and
the rest from the second machine, what is the
largest possible value of n which gives a
probability of at least 0.95 of the total mass of
the bags exceeding 240 kg?
(AEB)

20