4 Continuous Random

Variables and Probability

Distributions

Chapter 4 - Continuous Random Variables 1

 LO
Describe the probability distribution and calculate the mean and
variance of a continuous random variable
Continuous Uniform Distribution
Normal Random Variable
Exponential Distribution

Chapter 4 - Continuous Random Variables 2

Continuous Random Variables
connection
speed
length
Continuous or size
variables can take Continuous
physical
r. v.
any value of an examples
measurements

interval in 
height
durations or weight
of time

reaction time lifetime

waiting time
Chapter 4 - Continuous Random Variables 3
 Probability Density Function (pdf)
“density” is a measure of
f(x)  0 “probability mass” per unit length


Probability determined from the area under f(x). pdf
න 𝑓 𝑥 𝑑𝑥 = 1
f(x)
−
Ex. Let X be a continuous random variable and
suppose that f(x) = cx2, for -1 < x < 2, and 0 otherwise
𝑏
is the pdf of X. P(a  X  b) = ‫𝑓 ׬‬ 𝑥 𝑑𝑥
𝑎
a/ Find c
b/ Find P(X  0) = P(a  X < b)
------ = P(a < X  b) P(X = x0) = 0
 2 (different from pmf)
a/ ‫׬‬− 𝑓 𝑥 𝑑𝑥 = ‫׬‬−1 cx2𝑑𝑥 = c(8/3 +1/3) = 3c = P(a < X < b)

f(x) is the pdf of X  ‫׬‬− 𝑓 𝑥 𝑑𝑥 = 1  c = 1/3
2
b/ P(X  0) = ‫׬‬0 (1/3)x2𝑑𝑥 = 8/9
Chapter 4 - Continuous Random Variables 4
 Probability Density Function Pdf – Ex
The lifetime, in years, of some electronic component is a
continuous random variable with the probability density function
𝑘
𝑓 𝑥 = ቐ 𝑥3 𝑓𝑜𝑟 𝑥 > 1
0 𝑜𝑡ℎ𝑒𝑟𝑤𝑖𝑠𝑒
Find k and compute the probability for the lifetime to exceed 5
years.
  3
• ‫׬‬− 𝑓 𝑥 𝑑𝑥 = ‫׬‬1 k/x 𝑑𝑥 = -k(0 – 1)/2 = 1  k = 2

Chapter 4 - Continuous Random Variables 5

Cumulative Distribution Function (cdf)
F(x) = P(X  x)
= P(X < x) F′(x) = f(x)

x
P(a  X  b) cdf
F(x) = න 𝑓 𝑢 𝑑𝑢 Pdf: f(x) = x2/3, for -1 < x < 2
= F(b) – F(a) F(x)
−

F(x)  F(y) limx→ - F(x) = 0

if x  y limx→  F(x) = 1
(increasing)
x
Cdf: F(x) = ‫׬‬−1 t2/3𝑑𝑡 = x3/9 + 1/9
for -1 < x < 2, and 0 otherwise
Chapter 4 - Continuous Random Variables 6
 Pdf vs Cdf – Ex
F(x) = P(X  x) Ex. The lifetime, in years, of some
electronic component is a continuous
= P(X < x) F′(x) = f(x) random variable with the probability
density function
x
P(a  X  b) cdf 2
F(x) = න 𝑓 𝑢 𝑑𝑢 𝑓 𝑥 = ቐ 𝑥3 𝑓𝑜𝑟 𝑥 > 1
= F(b) – F(a) F(x)
− 0 𝑜𝑡ℎ𝑒𝑟𝑤𝑖𝑠𝑒
Find the cdf F(x) and P(2 < X < 3).
F(x)  F(y) 𝑥 𝑥 𝑥 2
limx→ - F(x) = 0 F(x) = ‫׬‬− 𝑓 𝑢 𝑑𝑢 = ‫׬‬1 𝑓 𝑢 𝑑𝑢 = ‫׬‬1 𝑑𝑢
if x  y 𝑢3
limx→  F(x) = 1 = 1 – 1/x2 for x > 1
(increasing)
P(2 < X < 3) = F(3) – F(2)
= ¼ - 1/9 = 0.139

Chapter 4 - Continuous Random Variables 7

 Mean and Variance
Expected Measure
value of X central
value
Expected
value of g(x) Mean  = E(X)
E[g(x)]  ∞
= ‫׬‬−∞ 𝑥𝑓 𝑥 𝑑𝑥
∞
= ‫׬‬−∞ 𝑔(𝑥)𝑓 𝑥 𝑑𝑥

Ex. Given the pdf of X,

X
Standard 2 = V(X)
3
𝑓 𝑥 = ቐ 𝑥4 𝑓𝑜𝑟 𝑥 > 1
Variance
∞
deviation 2 = ‫׬‬−∞ 𝑥 −  𝑓 𝑥 𝑑𝑥
2
 0 𝑜𝑡ℎ𝑒𝑟𝑤𝑖𝑠𝑒
Find the mean and variance of X.
 = 2
Measure spread 2 = E(X2) – E(X)2 ∞ ∞
of values E(X) = ‫׬‬−∞ 𝑥𝑓 𝑥 𝑑𝑥 = ‫׬‬1 3/𝑥3𝑑𝑥 = 1.5
V(X) = E(X2) – E(X)2 = 0.75

Chapter 4 - Continuous Random Variables 8

Chapter 4 - Continuous Random Variables 9
 Continuous Uniform Distribution
used for computer used in any situation when a i.e., locations of errors
simulation of various value is picked “at random” from in a program, birthdays
events and processes a given interval throughout a year

Variance: Uniform
pdf:
2 = (b – a)2/12 Distribution f(x) = 1/(b-a)

For h > 0,
Mean: P(t < X < t + h) = h/(b-a)
 = (a + b)/2 is independent of t

Ex. Tri is taking a bus from Dist.1 to Thu Duc city, a distance of 15km. His position is uniformly
distributed between the two areas. What is the probability that he is past Saigon Bridge, which
is 5km from Dist.1?
Let X be Tri’s location ➔ X ∼ Unif(0, 15) ➔ P(X > 5) = (5 < X < 15) = (15 – 5)/(15 – 0) = 2/3.

Chapter 4 - Continuous Random Variables 10

 Continuous Uniform Distribution – Ex
(Uniform Current) Let the continuous random variable X denote
the current measured in a thin copper wire in milliamperes.
Assume that the range of X is [0, 20 mA], and assume that the
probability density function of X is f(x) = 0.05, 0  x  20. Find
P(5 < X < 10), E(X), V(X).
-----
P(5 < X < 10) = (10 – 5)(0.05) = 0.25
E(X) = (a + b)/2 = (0 + 20)/2 = 10
V(X) = (b – a)2/12 = (20 – 0)2/12 = 33.33

Chapter 4 - Continuous Random Variables 11

 Normal Distribution (Gaussian
Plays a vital role in
distribution)
Probability and Statistics Good model for physical variables like weight, height,
temperature, voltage, pollution level (household
incomes or student grades, etc.)

Normal
N(, 2) distribution Related to the Central Limit Theorem (chapter 7)

Bell-shaped curve

Not easy to compute Normal probability density functions

with paper and pencil for selected values of  and 2.

Chapter 4 - Continuous Random Variables 12

 Normal Distribution
For any normal random variable

Probability that X > 13 for a normal random

variable with and  = 10 and 2 = 4.

6 is often referred to as the

width of a normal distribution.

Chapter 4 - Continuous Random Variables 13

 Standard Normal Random Variable
Standardize: If X ~ N(, 2),
then Z = (X - )/ ~ N(0, 1) It is the key step to calculating
a probability for an arbitrary
normal random variable.

Standard Assume Z is a standard normal random variable.

Z ~ N(0, 1) normal
Appendix Table III provides probabilities of the form
distribution
(z) = P(Z  z). Read down the z column to the row
that equals 1.5. The probability P(Z  1.5) is read from
the adjacent column, labeled 0.00, to be 0.93319.
(z) = P(Z  z)

0.93319

Chapter 4 - Continuous Random Variables 14

 Using Table III in the text book
to find probabilities of Normal
distribution
(1.34) = P(Z  1.34) = 0.909877

Chapter 4 - Continuous Random Variables 15

Chapter 4 - Continuous Random Variables 16
 P(-1.25 < Z < 0.37) = P(Z < 0.37) – P(Z < -1.25)

(6) Find z such that

P(Z > z) = 0.05.
Equivalently, find z
such that
P(Z  z) = 0.95

(5) P(Z < -4.6) cannot be found exactly from (7) Find z such that
Appendix Table III. The last entry in the P(-z < Z < z) = 0.99.
table can be used to find that P(Z < -3.99) = ➔ Find z such that
0.00003. So, 0  P(Z < -4.6) < P(Z < -3.99). P(Z  z) = 0.995

Chapter 4 - Continuous Random Variables 17

Chapter 4 - Continuous Random Variables 18
 Standardizing a Normal Random
Variable - Ex
Suppose the current measurements in a strip of wire are assumed to follow a
normal distribution with a mean of 10 milliamperes and a variance of 4
(milliamperes)2. What is the probability that a measurement will exceed 13
milliamperes?
Let X denote the current in milliamperes.
Let Z = (X - 10)/2.
Note that X >13 corresponds to Z >1.5.
Therefore, from Appendix Table III
P(X > 13) = P(Z > 1.5)
= 1 – P(Z 1.5) = 1- 0.93319 = 0.06681

Practical Interpretation: Probabilities for any normal random variable can be computed with a simple
transform to a standard normal random variable.

Chapter 4 - Continuous Random Variables 19

 Standardizing to Calculate a
Probability
(Normally Distributed Current) Continuing the previous example.
a/ What is the probability that a current measurement is between
9 and 11 milliamperes?
b/ Determine the value for which the probability that a current
measurement is below this value is 0.98.

Chapter 4 - Continuous Random Variables 20

 Normal
Approximation
to the
Binomial
Distribution

Binom(n, p) ~ N( = np, 2 = np(1-p))

Good approximation with n and p such

that
np > 5, n(1 – p) > 5

Chapter 4 - Continuous Random Variables 21

 Normal Approximation to the Binomial
Distribution
• X ~ Binom(n, p)
X – np
•Z= ~ N(0, 1)
np(1 – p)
• Normal Approximation of Binomial as follow
x + 0.5 – np
P(X  x) = P(X  x + 0.5) = P( Z  )
continuity correction
np(1 – p)
x – 0.5 – np
P(x  X) = P(x – 0.5  X) = P( Z)
np(1 – p)

Chapter 4 - Continuous Random Variables 22

Normal Approximation to the Binomial Distribution – Ex
Assume that in a digital communication channel, the number of bits
received in error can be modeled by a binomial random variable, and
assume that the probability that a bit is received in error is 10-5. If 16
million bits are transmitted, what is the probability that 150 or fewer
errors occur?
Let the random variable X denote the number of errors. Then X is a
binomial random variable with n = 16106 and p = 10-5, and

DIFFICULT!
Use normal distribution to approximate

Chapter 4 - Continuous Random Variables 23

Ex. A new computer virus attacks a folder consisting of 200 files.
Each file gets damaged with probability 0.2 independently of
other files. What is the probability that fewer than 50 files get
damaged?

Chapter 4 - Continuous Random Variables 24

 Conditions for approximating
hypergeometric and binomial probabilities

N = 400, K = 120, n = 30 n = 30, p = 0.3  = np = 9, 2 = np(1-p) = 6.3

Chapter 4 - Continuous Random Variables 25

 Normal Approximation to the Poisson
Distribution
• X is a Poisson random variable with E(X) = λ and V(X) = λ
X–λ
•Z= ~ N(0, 1)
λ
• Normal Approximation to Poisson as follow
x + 0.5 – λ
P(X  x) = P(X  x + 0.5) = P( Z  )
λ
x – 0.5 – λ
P(x  X) = P(x – 0.5  X) = P( Z)
λ

Chapter 4 - Continuous Random Variables 26

 Normal Approximation to the Poisson
Distribution – Ex
Ex. Assume that the number of asbestos particles in a squared
meter of dust on a surface follows a Poisson distribution with a
mean of 1000. If a squared meter of dust is analyzed, what is the
probability that 950 or fewer particles are found?
Let X be the number of particles found. Then X ~ Pois(1000)

DIFFICULT!
AN APPROXIMATION

Chapter 4 - Continuous Random Variables 27

 Exponential Distribution
• • • •

X = waiting time for the next event

Consider the sequence of “rare” events, where the number of occurrences

has Poisson distribution with parameter λ.
Event “the time X until the next event is greater than x” can be rephrased as
“zero events occur by the time x.”
Let N be the number of events during the time interval [0, x] ➔
N~Poisson(λx).
P(X > x) = P(N = 0) = e-λx(λx)0/0! = e-λx
➔ F(x) = P(X  x) = 1 P(X > x) = 1 – e-λx
➔ f(x) = F(x) = λe-λx
Chapter 4 - Continuous Random Variables 28
Exponential Distribution
Often used to model waiting time between rare events:
time between telephone calls, accidents, etc. When the
number of events is Poisson, the time between events is
Exponential.

Exp.
distribution

For small lambda,

it’s usual that we
need a long
waiting time for the
next event

Chapter 4 - Continuous Random Variables 29

 Exponential Distribution – Ex
Ex. Questions are sent to the chatbot of a Student Service at an
average rate of 12 questions per day.
a/ What is the probability that the next question is sent within 1 hour
minutes?
b/ What is the expected time between questions?
Questions arrivals represent rare events, thus the time X between
them has an exponential distribution with λ = 12 (questions per day).
a/ 1hr = 1/24 day, P(X  1/24) = F(1/24) = 1 − e−λx = 1 − e−1/2 = 0.3935
b/ E(X) = 1/λ = 1/12 (day) = 2 hours

Chapter 4 - Continuous Random Variables 30

 Exponential Distribution – Ex
Suppose that X has an exponential distribution with a mean of
10. Determine the following:
a/ P(X < 5)
b/ P(X < 15 | X > 10)
E(X) = 1/ = 10 ➔  = 1/10
a/ P(X < 5) = F(5) = 1 − e−5λ = 1 − e−1/2 = 0.3935
b/ P(X < 15 | X > 10) = P(10 < X < 15)/P(X > 10)
= (F(15) – F(10))/(1 – F(10)) = [(1 − e−15λ ) – (1 − e−10λ)]/e−10λ
= e−10λ(1 - e−5λ)/e−10λ = 1 - e−5λ = 0.3935 = P(X < 5)

Chapter 4 - Continuous Random Variables 31

 Lack of Memory Property
For an exponential random variable X,
P(X < t1 + t2 | X > t1) = P(X < t2)
Ex. The time between the arrival of electronic
messages at your computer is exponentially
distributed with a mean of two hours ( = 1/2).
a/ What is the probability that you do not receive a
message during a two-hour period?
P(X > 2) = 1 – P(X  2) = e-2 = e-1 = 0.368
b/ If you have not had a message in the last four
hours, what is the probability that you do not
receive a message in the next two hours?
P(X > 6 | X > 4) = P(X > 2)

Chapter 4 - Continuous Random Variables 32

 Exercises
(5x + 19) mod 109
(7x + 29) mod 109
(9x + 39) mod 109

Chapter 4 - Continuous Random Variables 33

 SUMMARY

Continuous
Uniform

Continuous
variables can take Continuous
any value of an r. v.
Normal
interval in 

Exponential

Chapter 4 - Continuous Random Variables 34