DPBS 1203

Business and Economic Statistics

Lecture 3.2
Discrete vs. continuous distributions

 Random variable — a function that assigns a real

number to each possible value in the sample space of
an event
 Discrete random variable — is the one which takes a
finite or countably infinite number of values:
– Number of new subscribers to a magazine.
– Number of absent employees on a given day.
Discrete vs. continuous distributions

 Continuous random variable — takes all

possible values in some interval:
– Elapsed time between arrivals of bank customers.
– Percent of the labour force that is unemployed.
Types of distributions

 Discrete  Continuous
– binomial – uniform
– Poisson – normal
– exponential
– t
– chi-square
– F
The binomial distribution
 Based on the notion of a “binomial experiment” consisting of a
sequence of trials
 Requirements/assumptions for something to qualify as a binomial
experiment:
– Sequence of fixed number of n trials
– Each trial has 2 outcomes, arbitrarily denoted “success” and “failure”
– There is a fixed probability of success (failure) p (1-p) over all trials
– Trials are independent
– Under these assumptions, this is a sequence of Bernoulli trials
– The outcome of each trial is recorded in a random variable
Binomial random variables
 Represent our sequence of Bernoulli random variables:
– X1, X2, …, Xn where Xi = 1 is called “success” and Xi = 0 is called
“failure”
– Under the assumptions made, this is a sequence of
independent and identically distributed (iid) random variables
 Suppose n = 5. Then:
– One possible outcome of the sequence of trials is 0, 1, 1, 0, 1 .
– Another is 1, 1, 0, 0, 0. And so on.
Binomial random variables…
 We can form another random variable from these Bernoulli
random variables
– Consider the random variable formed by summing these n
Bernoulli random variables:
X = X1 + X2 + … + Xn
 X represents the number of successes in n trials (note that
order does not matter!)
 X is called a binomial random variable
– Characterized by 2 parameters: n and p
– i.e. once we know the values of these parameters in a given
case, we know everything about the random variable and its
probability distribution
Binomial distribution
 In the gender composition example, we can calculate:
P(MMF) = P(X1=1, X2=1, X3=0)= pp(1-p) = p2(1-p)

Now, there are 3 ways to obtain 2 males (the two others being MFM or
FMM), so
P(2 Males) = P(X = 2) = 3p2(1-p)

 The “3” coefficient on the expression for the probability happens to

correspond to the total number of possible combinations when
“choosing 2 from 3”.
Binomial distribution
 Recall counting techniques (combinations and permutations)?
 Here, what is relevant is the combinatorial formula: i.e. how many distinct
combinations exist when we choose 𝑥𝑥 objects from 𝑛𝑛 objects, without regard
to order?
𝑛𝑛!
 𝐶𝐶𝑥𝑥𝑛𝑛 = where 𝑛𝑛! = 𝑛𝑛(𝑛𝑛 − 1)(𝑛𝑛 − 2). . . (2)(1) and 0! = 1
𝑥𝑥!(𝑛𝑛−𝑥𝑥)!
3!
 If 𝑛𝑛 = 3 & 𝑥𝑥 = 2 then 𝐶𝐶23 = =3
2!(3−2)!
Binomial distribution…
The probability of observing x successes in a binomial experiment with n trials
and probability of success p is
n!
P( X = x) ≡ P( x) = p x (1 − p ) ( n − x ) for x = 0,1,  , n
x!(n − x)!

 Members of this family of distributions are characterized by n and p

 Different binomial random variables are generated with different n
and p combinations
 Given a particular set of values for n and p , it is irrelevant whether
we’re considering a coin toss, family composition, or anything else.
The process follows the same distribution.
Binomial example
n!
P( X = x) ≡ P( x) = p x (1 − p ) ( n − x ) for x = 0,1,  , n
x!(n − x)!

 A multiple choice exam has 10 Here n = 10, p = 0.2, and it is reasonable to assume
questions, each of which has 5 choices that the number of correct answers is a binomial
of which only one is correct. random variable.
 If a student uses merely random P ( X = 5)=10 C5 (.2) 5 (.8) 5
guesswork to answer the quiz, what is: 10!
= (.2) 5 (.8) 5 = .0264
– P(X = 5)? 5!5!
– P(X ≤ 1)? P ( X ≤ 1) = P ( X = 0) + P ( X = 1)
– P(X > 1)? 10! 10!
= (.2) 0 (.8)10 + (.2)(.8) 9
0!10! 1!9!
= .1047 + .2684 = .3758
P ( X > 1) = 1 − P ( X ≤ 1) = 1 − .3758
Binomial example…
 In this example
– P(X = 5) is a binomial probability
– P(X ≤ 1) is a cumulative binomial probability
– P(X > 1) is sometimes called a survivor probability
 Even with counting techniques, these probabilities can be cumbersome
to evaluate
 So, values of binomial probabilities have also been tabulated
– Tables exist of both individual probabilities and cumulative probabilities
Using binomial tables
 For P(X=5) when n=10, p=0.2:
– We need the column labelled “p=0.2” in the block of values
corresponding to “n=10”
– Now consider the row labelled “k=5”  0.0264
 P(X ≤ 1) could be verified from a table of individual probabilities by
summing the relevant ones
– Or directly from a table of cumulative probabilities
Binomial tables…
Binomial Probability P(X=k)
p
n k 0.05 0.10 0.15 0.20 0.25 0.90 0.95
2 0 0.9025 0.8100 0.7225 0.6400 0.5625 0.0100 0.0025
1 0.0950 0.1800 0.225 0.3220 0.3750 0.1800 0.0950
2 0.0025 0.0100 0.0225 0.0400 0.0625 0.8100 0.9025

3 0 0.8574 0.7290 0.6141 0.5120 0.4219 0.0010 0.0001

1 0.1354 0.2430 0.3251 0.3840 0.4219 0.0270 0.0071
2 0.0071 0.0270 0.0574 0.0960 0.1406 0.2430 0.1354
3 0.0001 0.0010 0.0034 0.0080 0.0156 0.7290 0.8574

10 0 0.5987 0.3487 0.1969 0.1074 0.0563 0.0000 0.0000

1 0.3151 0.3874 0.3474 0.2684 0.1877 0.0000 0.0000
2 0.0746 0.1937 0.2759 0.3020 0.2816 0.0000 0.0000
3 0.0105 0.0574 0.1298 0.2013 0.2503 0.0000 0.0000
4 0.0010 0.0112 0.0401 0.0881 0.1460 0.0001 0.0000
5 0.0001 0.0015 0.0085 0.0264 0.0584 0.0015 0.0001
6 0.0000 0.0001 0.0012 0.0055 0.0162 0.0112 0.0010
7 0.0000 0.0000 0.0001 0.0008 0.0031 0.0574 0.0105
8 0.0000 0.0000 0.0000 0.0001 0.0004 0.1937 0.0746
9 0.0000 0.0000 0.0000 0.0000 0.0000 0.3874 0.3151
10 0.0000 0.0000 0.0000 0.0000 0.0000 0.3487 0.5987
 Binomial tables…
Binomial distribution with n = 10, p = 0.2

 Tables provide the entire 0.40

distribution 0.30

– What does this distribution look like

Probability
0.20

for n = 10, p = 0.2? 0.10

 Suppose that with study, the chance 0.00

of correctly answering a given 0 1 2 3 4 5

Correct answers
6 7 8 9 10

question improves to p = 0.5. Binomial distribution with n = 10, p = 0.5

– What is the binomial distribution in 0.30

this case? 0.25

– What are the key features of these

0.20

Probability
0.15

2 distributions? 0.10

– What would you expect for p = 0.8?

0.05

0.00
0 1 2 3 4 5 6 7 8 9 10
Correct answers
Mean and variance of a binomial random variable
𝐸𝐸(𝑋𝑋𝑖𝑖 ) = 0 × (1 − 𝑝𝑝) + 1 × 𝑝𝑝 = 𝑝𝑝
𝑉𝑉𝑉𝑉𝑉𝑉(𝑋𝑋𝑖𝑖 ) = (0 − 𝑝𝑝)2 (1 − 𝑝𝑝) + (1 − 𝑝𝑝)2 𝑝𝑝
= 𝑝𝑝(1 − 𝑝𝑝)(𝑝𝑝 + 1 − 𝑝𝑝) = 𝑝𝑝(1 − 𝑝𝑝)

Thus, 𝑛𝑛 𝑛𝑛

𝐸𝐸(𝑋𝑋) = 𝐸𝐸(� 𝑋𝑋𝑖𝑖 ) = � 𝐸𝐸(𝑋𝑋𝑖𝑖 ) = 𝑛𝑛𝑛𝑛

𝑖𝑖=1 𝑛𝑛 𝑖𝑖=1 𝑛𝑛

𝑉𝑉𝑉𝑉𝑉𝑉(𝑋𝑋) = 𝑉𝑉𝑉𝑉𝑉𝑉(� 𝑋𝑋𝑖𝑖 ) = � 𝑉𝑉𝑉𝑉𝑉𝑉(𝑋𝑋𝑖𝑖 ) = 𝑛𝑛𝑛𝑛(1 − 𝑝𝑝)

𝑖𝑖=1 𝑖𝑖=1
Mean & variance of a binomial random variable…
x 0 1 2 3
P(X=x) 0.125 0.375 0.375 0.125

 Previously, we calculated the mean and variance of X in the

gender composition example from first principles
– Let’s apply the formula we just derived!
– E(X) = np = 3×0.5 = 1.5
– Var(X) = np(1-p) = 3×0.5×0.5 = 0.75
Models as approximations
 In the gender composition example:
– We defined X = number of boys in families with 3 children
– We assumed:
 Birth outcomes are independent events
 The probability of a boy being born does not vary over births
 Males and females are equally likely to be born
 The first 2 assumptions are required for “binomial” to be an appropriate
description of the distribution of X
– How reasonable is each of the three assumptions?
Other common discrete distributions
 There are many other common discrete distributions that can be used to
represent or model real phenomena
e.g. the discrete uniform distribution (pretty boring!)
 The Poisson distribution is a little more interesting
– Binomial – number of successes in sequence of trials
– Poisson – number of successes in a period of time or region of space
– The Poisson distribution relates to a count variable (0, 1, 2, …), where
successes are relatively rare
e.g. the number of times an individual visited a GP in the last year
Poisson distribution
 Probability Distribution

 The variance is 𝜆𝜆 the same with the mean and the standard
deviation is the square root of 𝜆𝜆.
Review
 Introduction to random variables
 Mathematical expectations
 Discrete random variables
 The binomial distribution (and Bernoulli trials), in depth
 Other discrete distributions, in brief