CHAPTER 15 : RANDOM VARIABLES

1. The discrete random variable X has the probability distribution as follows, 3. The number of car accidents at a particular junction is not more than four
x –60 20 24 60 accidents in a day. The probability that there is no car accident in a day is 0.4
P(X = x) p 0.10 q 0.25 and the probabilities that there are at most one, two and three car accidents in
a day are 0.7, 0.85 and 0.95 respectively.
It is given that P(X ≤ 20) = 0.25. (a) Construct a probability distribution table for the number of car accidents
(a) Determine the values of p and q. [2] at the junction in a day. [3]
2 (b) Calculate the mean of the number of car accidents at the junction in a day. [2]
(b) Find E(X) and E{[X – E(X)] }. [5]
(a) X = no. of car accidents at the junction in a day
(a) P(X ≤ 20) = 0.25 : p + 0.1 = 0.25  p = 0.15
P(X = 0) = 0.4 , P(X ≤ 1) = 0.7 , P(X ≤ 2) = 0.85 and P(X ≤ 3) = 0.95
 P(X = x) = 1 : (0.15) + 0.1 + q + 0.25 = 1  q = 0.5
x 0 1 2 3 4
(b) E(X) = (–60)(0.15) + 20(0.1) + 24(0.5) + 60(0.25) = 20 P(X = x) 0.40 0.30 0.15 0.10 0.05
2 2 2 2 2
E{[X – E(X)] } = E{[X – 20] } = (–80) (0.15) + 0 + (4) (0.5) + (40) (0.25) = 1368
Alternative
2 2 2 2 2 (b) Mean = E(X) = 0(0.4) + 1(0.3) + 2(0.15) + 3(0.1) + 4(0.05) = 1.1
E(X ) = (–60) (0.15) + (20) (0.1) + (24) (0.5) + (60) (0.25) = 1768
2 2 2
E{[X – E(X)] } = E{[X – 20] } = E(X ) – 40 E(X) + E(400)
= 1768 – 40(20) + 400 = 1368
4. On average, 48 cars enter a car park of a shopping complex per hour. The
number of cars which enter the car park has a Poisson distribution.
2. A discrete random variable X which takes the values 0, 1, 2 and 3, and its (a) Find the probability that at most two cars enter the car park in a five-minute
probability distribution is shown in the table below. interval. [3]
x 0 1 2 3 (b) Determine time interval, in minutes, if the probability that at least one car
2a enters the car park in this interval is 0.85. [4]
P(X = x) a b 0.15
3
(a) X = no. of cars entering in 5 min, X ~ Po  5 x 48   X ~ Po(4)
If E(X) = 1.20, find the values of a and b. [4]  60 
2
Hence, show that X is not a binomial random variable. [2] P(X ≤ 2) = P(X = 0) + P(X = 1) + P(X = 2) = e 4 (1 + 4 + 4 ) = 0.23810
2!
 P(X = x) = 1 : a + b + 2 a + 0.15 = 1  b + 5 a = 0.85
3 3
(b) T = no. of cars entering in t min. interval, T ~ Po  t x 48   T ~ Po( 4 t )
E(X) = 1.20 : 0 + b + 2( 2 a ) + 3(0.15) = 1.20  b + 4 a = 0.75  60  5
3 3
P(T ≥ 1) = 0.85  1 – P(T = 0) = 0.85
 a = 0.3 , b = 0.35
 4t
0.85 = 1 – e 5  – 4 t = ln (0.15)  t = 2.3714
Assume X ~ B(3, p), since given that E(X) = 1.20 : np = 3p = 1.2  p = 0.4 5
3
Since P(X = 0) = 0.6 = 0.216 ( ≠ a)  X is not a binomial random variable
Allternative :
3
Since P(X = 3) = 0.4 = 0.064 ( ≠ 0.15)  X is not a binomial random variable

15N-1
5. A laboratory equipment shop gets 45% of the electronic weighing scale supplies 7. The continuous random variable X
from country Q. Five electronic weighing scales are selected at random. has the probability density function  k ,0  x  1
(a) Determine the mean and standard deviation for the number of the electronic 4  k
(a) Show that k = . [3] f(x) =  , 1 x  4
weighing scales supplied by country Q. [4] 7 x2

(b) Calculate the mean and Var(X). [7]  0 , otherwise
(b) Find the probability that less than three electronic weighing scales are
supplied by country Q. [4] k
[kx ]01 + [ – k ]14 = 1
1 4
(a) P(–  < X < ) = 1 :  0 k dx +  1 x2
dx = 1 
x
X = no. of electronic weighing scales supplied by country Q, X  B(5 , 0.45)
(a) Mean = 5(0.45) = 2.25 , Standard deviation = 5(0.45)(0.55) = 1.1124 [ k – 0] – [ k – k ] = 1
4
 k=
4
7
(b) P(X < 3) = P(X = 0) + P(X = 1) + P(X = 2)
5 0 5 5 1 4 5 2 3
= C0.(0.45) (0.55) + C1.(0.45) (0.55) + C2.(0.45) (0.55) = 0.59313 1 4 k 2 2 1 4
 0 kx dx +  1
4
(b) E(X) = dx = [ x ]0 + [ ln x ]1
x 7 7
2
6. Studies conducted in the past show that the mean number of serious accidents = [1 – 0] + 4 [ln 4 – 0] = 2 + 4 ln 4
in a certain factory is one per year. The factory wishes to find the probability of 7 7 7 7
the number of serious accidents occurring on a yearly basis for the purpose of 1 4 4 3 1 4
 
4
k dx = [ x ]0 + [ x ]1
2 2
revising the insurance premium of its employees. The factory is required to
E(X ) = kx dx +
0 1 21 7
increase the insurance premium if the probability of having less than two
4 4 40
serious accident-free years out of five years is more than 0.35. =
21
[ 1 – 0] + [4 – 0] =
7 21
(a) Define the relevant random variable, and identify its distribution. [3]
40 2 4 2 268 16 16
(b) Find the probability that
 Var(X) = –( + ln 4 ) = – ln 4 – (ln 4)2 or 0.7429
21 7 7 147 49 49
(i) there is no serious accident occurring in the next one year, [2]
(ii) there are at least three serious accidents occurring in the next two years, [4]
(iii) there are less than two accident-free years occurring out of five years. [5]
(c) Based on the result in (b)(iii), state whether the factory is required to
increase the insurance premium or not. [1]

(a) X = number of serious accidents in a certain factory in a year, X ~ Po(1)

–1
(b) (i) P(X = 0) = e = 0.3679
(ii) Y = number of serious accidents in a certain factory in 2 year, Y ~ Po(2)
–2 22
P(Y ≥ 3) = 1 – P(Y = 0, 1, 2) = 1 – e (1 + 2 + ) = 0.3233
2!
(iii) W = number of accident-free years, W ~ B(5, 0.3679)
5 0 5 5 4
P(W < 2) = P(W = 0, 1) = C0(0.3679) (0.6321) + C1(0.3679)(0.6321) = 0.3946

(c) Since P(W < 2) > 0.35, the factory is required to increase the insurance premium.

15N-2
8. The cumulative distribution function for the continuous random variable X is  0 , x 1
given by 1 2
 ,x  0  x  x  c1 ,1  x  2
0 (a) Calculate the median. [4] (b) F(x) =  2 1
 1 2 1  2 x  c2 ,2  x  3
 x ,0  x  1 (b) Find P(|X – 1| < ). [2] 
F(x) =  3 2  1 ,3  x
1 2 1
 x x  ,1  x  3 (c) Determine the probability
 6 2 density function of X. [2] x = 1: 0 = 1 – 1 + c1  c1 = 1 x=3: 1 (3) + c = 1
2  c2 = – 1
2 2 2 2
 1 ,x  3
 0 , x 1
1 2
(a) F(m) = 0.5  1 ≤ m < 3  ( x  2 x  1) ,1  x  2
 F(x) =  2 1
1 2 1 2  2 ( x  1) ,2  x  3
m – m – = 0.5  m – 6m + 6 = 0 
6 2  1 ,3  x
m=3– 3,3+ 3  (since 1 ≤ m < 3) : m = 3 – 3 Alternative (b)
1 (b) 1 < x ≤ 2 :
(b) P(|X – 1| < ) = P(0.5 < X < 1.5) = (1.5) – 61 (1.5)2 – 21
[ ] –
1 2
(0.5) =
13
x

2 3 24 1 2 x 1 1 1 2
F(x) = ( x – 1) dx = [ x – x ]1 = [ x 2 – x] – [ – 1] = (x – 2x + 1)
 2 1 2 2 2 2
 3x ,0 x 1
2 x 1
  [ 21 x] 2x
 1
d 2<x≤3: F(x) = ( x – 1) dx + dx = (4 – 4 + 1) +
(c) f(x) =
dx
[F(x)] =  1  1 x , 1  x  3 1 2 2 2
 3
 0 1 1 1
, otherwise = + [ x – 2] = ( x – 1)
 2 2 2
 0 , x 1
1 2
 ( x  2 x  1) ,1  x  2
 F(x) =  2 1
 2 ( x  1) ,2  x  3
9. The continuous random variable X has
probability density function  x  1 ,1  x  2
  ,3  x
f(x) =  k ,2  x  3  1
(a) Show that k = 1 . 0 [2]
2  , otherwise
(b) Find the cumulative distribution function. [4] (c) F(x)
(c) Sketch the graph of the cumulative distribution function. [2] 1
(d) Find P( 3 ≤ X ≤ 5 ). [2]
2 2 1
2
1
(a) (1)(0 + 1) + k(1) = 1  k=1
2 2
Alternative (a) 0 1 2 3 x
2 3
 ( x – 1) dx + 
1
k dx = 1 
2
[(x – 1)2 ]21 + [kx]32 = 1
1 2
(d) P( 3 ≤ X ≤ 5 ) = P(X ≤ 5 ) – P(X < 3 ) = 1 [ 5 – 1] – 1 [( 3 )2 – 2( 3 ) + 1] = 5
1 2 2 2 2 2 2 2 2 2 8
 (1 – 0) + k(3 – 2) = 1  k=1
2 2

15N-3
10. A continuous random variable X has probability  0 ,x  0
 k ,0  x  2
density function  x
 m   c1 ,0  x  2
It is given that 3 P(X ≤ 2) = 2 P(X > 2). f(x) =  ,2  x  5 (b) F(x) =  5
 x2 2
   c2 ,2  x  5
(a) Show that k = 1 and find the value of m. [5]  0 ,5  x
 x
5
 1 ,x  5
(b) Find the cumulative distribution function of X. [5]
 F(x) is continuous, 0 = 1 (0) + c1  c1 = 0
(c) Determine P(1 ≤ X ≤ 3). [2] 5
1 (2) + 0 = – 2 + c2  c2 = 7
(d) Calculate the median value of X. [3] 5 2 5
 0 ,x  0
(a) 3P(X ≤ 2) = 2 P(X > 2) = 2[1 – P(X ≤ 2)]  P(X ≤ 2) = 2  x
5
 ,0  x  2
F(x) =  5
2k = 2  k = 1 7  2 ,2  x  5
5 5 
 5 x
5
 1 ,x  5
5
2 +  m  = 1
P(–  < X < ) = 1 : 2 x 1 +
5 2 mx  2 dx = 1 
5  x  2
Alternative (b)
–m +m =3  m=2 x x
1 dx =  x  = 1 x
5 2 5 P(X ≤ x) = 0 5  5  0 5
Alternative (a)
x x
2 +  2  = 7 – 2

2
k dx + 
5
m dx = 1 , 3

2
k dx = 2 
5
m dx P(X ≤ x) = 2 +
5  2 x22 dx = 5  x  2 5 x
0 2 x2 0 2 x2
 0 ,x  0
kx  20 +  mx    20
5 5
 x
=1 3 kx = 2  m   ,0  x  2
 x  2
2 F(x) =  5
7 2
  ,2  x  5
(2k – 0 ) + (– m + m ) = 1 3(2k – 0) = 2(– m + m )  5 x
5 2 5 2  1 ,x  5
 k=1, m=2
5
(c) P(1 ≤ X ≤ 3) = P(X ≤ 3) – P(X < 1) = ( 7 – 2 )– 1= 8
5 3 5 15

(d) Since P(X ≤ m) = 1 , 2 < m ≤ 5

2
7– 2 = 1  m = 20
5 m 2 9

15N-4
11. A continuous random variable X has  2(k  x ) 12. (a) Describe briefly the standard normal random variable. [2]
a probability density function,  ,0  x  k
f(x) =  k2 (b) The life span of a certain light bulb is a normal random variable with a mean
 0 , otherwise of 950 hours and a standard deviation of 50 hours.
where k is a positive constant
(a) Find the mean of X in terms of k. [3] (i) Find the probability that a randomly chosen light bulb has a life span
k of more than 1000 hours. [3]
(b) Show that the standard deviation of X is equal to . [4]
3 2 (ii) Determine the value of h such that 99% of the light bulbs have life
span between (950 – h) hours and (950 + h) hours. [5]
(c) Find P(X <  +  2 ), where  and  are the mean and standard deviation
of X respectively. [5] (iii) Find the probability that at most one out of eight independently
selected light bulbs has a life span of more than 1000 hours. [5]
k
k 2(k  x ) 1  2x3  1  3 2k 3 

1
(a) Mean = x. dx = 2  kx 2   = k   –0= k
0 k2 k  3 
0
2
k  3  3 (a) Standard normal random variable is a continuous random variable
and with a mean zero and variance 1.

k
k 2(k  x ) 2  kx 3 x4  2  k4 k4 

2 1 2
(b) E(X ) = x2. 2
dx = 2
   =    –0= k 2
2
0 k k  3 4  0
k  3 4  6 (b) (i) Let X be the life span of a certain light bulb, X ~ N(950, 50 )
2 1000  950
1 2 1  k P(X > 1000) = P(Z > ) = P(Z > 1) = 0.1587
Variance of X = k –  k  Standard deviation of X = 50
6 3  3 2

h h
(ii) P(950 – h < X < 950 + h) = 0.99  P(– <Z< ) = 0.99
 2kx  x 
2 2
k 2( k  x )

2 3 1 k 50 50
(c) P(X <  +  2 ) = P(X < k ) = dx = 2 3
3 k2 k2 h h h
 P(Z <  h = 128.8
0
0 ) = 0.995 or P(Z > ) = 0.005  = 2.576
50 50 50
1  4k 2 4k 2  8
=
2
   –0= Note : Do not accept confidence interval method, cause population mean is known.
k  3 9  9

(iii) Let Y = no. of light bulbs that (X > 1000), Y ~ B(8, 0.1587)
8 7
P(Y ≤ 1) = P(Y = 0) + P(Y = 1) = (0.8413) + 8(0.1587)(0.8413) = 0.6299

15N-5
13. The masses of guavas in a farm are normally distributed with mean, , and 14. It is known that 2% of printers sold by a supplier malfunction within one year.
standard deviation, . The mass percentages of guava that less than 400 g The supplier offers a warranty period of one year.
is 15.87% and more than 500 g is 6.68%. (a) Find the probability that, out of 20 printers, at least three printers malfunction
(a) Determine the values of , and . [7] within the warranty period. [4]

(b) Find the probability that a guava has a mass of at most 500 g given that it is (b) Determine the least number of printers, such that the probability that at least
more than 400 g. [3] one printer malfunctions within the warranty period is more than 0.9995. [5]

An agriculture officer thinks that the number of guavas, Y, on a branch of (c) Using a suitable approximation, find the probability that, out of 1000 printers,
guava trees in the farm may be modelled by a Poisson distribution. there are between 10 to 18 printers (inclusive) malfunction within the warranty
period. [6]
(c) If P(Y = 0) = 0.0954, find the parameter of the Poisson distribution, [3]
(a) Let X = number of printers malfunction within the warranty period,
(d) Give two reasons why Poisson distribution cannot be an exact model for Y. [2]
X  B(20, 0.02)
(a) X = mass of guava : 2
(0.02) ( (0.98)20 x = 0.0070687
x
400  
P(X ≥ 3) = 1 – P(X = 0, 1, 2) = 1 –  20 C x
P(X < 400) = 0.1587 : = –1  400 –  = – ---- (1) x 0

500   (b) Let Y = number of printers malfunction within the warranty period,
P(X > 500) = 0.0668 : = 1.5  500 –  = 1.5 ---- (2)
 Y  B(n, 0.02)
  = 440 ,  = 40 P(Y ≥ 1) > 0.9995 : P(Y = 0) < 0.0005  0.98 < 0.0005
n

n log 0.98 < log 0.0005  n > 376.23

(b) P(X ≤ 500 | X > 400) =
P( 400  X  500) (1  0.0668 )  0.1587
= = 0.92060  Least number of printers required = 377
P( 400  X ) 1  0.1587

np  20 , nq  980
(c) Y  Po() , P(Y = 0) = 0.0954 : e
–
= 0.0954 (c) T  B(1000, 0.02)  T  N(20, 19.6)

  = – ln 0.0954 = 2.350 P(10 ≤ T ≤ 18) → P(9.5 ≤ T ≤ 18.5)

9.5  20 18.5  20
= P( ≤Z≤ ) = P(–2.372 ≤ Z ≤ –0.339)
(d) R1 : The space (a branch of tree) defined in Y 19.6 19.6
is not a constant /consistent(not specific)
= P(Z ≤ 2.372) – P(Z ≤ 0.339) = P(–0.339) – P(–2.372) = R(–2.372) – R(–0.339)
R2 : The size of guavas taken into account is not specified. = 0.9912 – 0.6327 = 0.3673 – 0.0088 = 0.9912 – 0.6327
= 0.3585 = 0.3585 = 0.3585

15N-6
15. In a given population, 25% of the population choose jogging as their leisure (b) Use normal approximation to binomial
activity. 2
 = 60(0.25) = 15 = 60(0.25 )(0.75 ) = 3.3451 or  = 11.25
(a) Find the probability that, of the 20 people selected at random,
(i) exactly five people choose jogging as their leisure activity, [3] P(12 ≤ X ≤ 15) → P(11.5 ≤ X ≤ 15.5)
 11.5  15 15.5  15 
(ii) at most 17 people do not choose jogging as their leisure activity. [4] = P Z  = P(–1.044 ≤ Z ≤ 0.149)
 3.3451 3.3451 
(b) Use a suitable approximation to find the probability that, out of 60 selected at
random, 12, 13, 14 or 15 choose jogging as their leisure activity. Justify the = P(Z ≤ 0.149) – P(Z ≤ –1.044) = P(0.149) – P(–1.044) = R(–1.044) – R(0.149)
choice of your approximation. [8] = 0.5592 – (1 – 0.8518) = 0.5592 – 0.1482 = 0.8518 – 0.4408
= 0.4110 = 0.4110 = 0.4110
(a) (i) X = No. of people who choose jogging as their leisure activity
Justification : np = 15 (> 5) and nq = 45 (> 5)
X  B(20 , 0.25)
20 5 15
P(X = 5) = C5 . (0.25) (0.75) = 0.20233
16. The discrete random variable X has probability function
(ii) P(X ≥ 3) P(X ≥ 3) = 1 – P(X < 3)  k ( 4  x )2 , x  1, 2, 3

= 1 – [P(X = 0) + P(X = 1) + P(X = 2)] = P3 + P4 + P5 + . . . + P20 P(X = x) =  , where k is a constant.

 0 , otherwise
2 20
 
20 x 20 – x 20 x 20 – x
=1– Cx . (0.25) (0.75) = Cx . (0.25) (0.75) (a) Determine the value of k and tabulate the probability distribution of X. [3]
x 0 x3
(b) Find E(7X – 1) and Var(7X – 1). [7]
= 0.90874 = 0.90874
1
Alternative (a)  P(X = x) = 1 : k(3)2 + k(2)2 + k(1)2 = 1  k =
(iI) Y = No. of people who do not choose jogging as their leisure activity 14
x 1 2 3
Y  B(20 , 0.75)
9 4 1
P(Y ≤ 17) P(Y ≤ 17) = 1 – P(Y > 17) P(X = x)
14 14 14
= P0 + P1 + P2 + . . . + P17 = 1 – [P(Y = 18) + P(Y = 19) + P(Y = 20)]
17 20
 
20 y 20 – y 20 y 20 – y
= Cy . (0.75) (0.25) =1– Cy . (0.75) (0.25)
9 4 1
y0 y  18 (b) E(7X – 1) =  (7x – 1).P(X = x) = 6( ) + 13( ) + 20( )=9
14 14 14
= 0.90874 = 0.90874
Var(7X – 1) = E[(7X – 1) ] – [E(7X – 1)]
2 2

9 4 1
= [(6) .( )] – (9)2 = 19
2
)+ (13)2.( )+ (20)2.(
14 14 14

15N-7