By:

Prof. Mahmoud Gabr

Dr Nesrin Nabil Dr Marwa Hassan
Chapters
1. Bivariate (Joint) Distributions

2. Sampling Distributions
3. Point Estimation

4. Interval Estimation

5. Tests of Hypotheses

6. Analysis of Variance (Anova)

Chapter 1:
Bivariate (Joint) Distributions
Revision
 Random Variable
Discrete RV Continuous RV

Countable (finite or infinite) Uncountable (any value in an interval)

Defined at specific points Defined over intervals

as: 𝒙=1,2,…,10 or 𝒙 = 𝟎, 𝟏, 𝟐, … as: 𝒙 ≥ 𝟎 or 𝟓 < 𝒙 < 𝟏𝟎

Number of heads in 10 flips Time between bus arrivals

probability mass function (p . m. f . ), Probability Density Function (p.d.f),

𝒃
𝒇 𝒙 =𝑷 𝑿=𝒙 , 𝒙 = 𝒙𝟏 , 𝒙𝟐 , … 𝑷 𝒂≤𝑿≤𝒃 = ‫𝒇 𝒂׬‬ 𝒙 𝒅𝒙
 Discrete Random Variable

• If 𝑿 is a discrete r.v. with possible values 𝒙𝟏 , 𝒙𝟐 , … , then

𝒇 𝒙 =𝑷 𝑿=𝒙 , 𝒙 = 𝒙𝟏 , 𝒙𝟐 , …
is called the probability mass function (p.m.f.) of 𝑿.
• The p.m.f. must satisfies the following conditions
1. 𝒇 𝒙 ≥ 𝟎 for all 𝒙
2. σ𝒂𝒍𝒍 𝒙 𝒇 𝒙 = 𝟏
 Discrete Random Variable
Example:
• Let's define 𝑿 as the number of heads when a fair coin is tossed twice.

• When tossing a fair coin twice, the possible outcomes are: HH,HT,TH,TT
• Now, find 𝑷(𝑿 = 𝒙) for each possible value of 𝑿:
1. 𝒇 𝟎 = 𝑷(𝑿 = 𝟎) = 𝑷(𝑻𝑻) = 𝟏/𝟒
2. 𝒇 𝟏 = 𝑷 𝑿 = 𝟏 = 𝑷 𝑯𝑻 + 𝑷 𝑻𝑯 = 𝟏Τ𝟒 + 𝟏Τ𝟒 = 𝟐Τ𝟒
3. 𝒇 𝟐 = 𝑷 𝑿 = 𝟐 = 𝑷 𝑯𝑯 = 𝟏Τ𝟒
𝒙 0 1 2 Sum
• Thus, the Probability Mass Function (p.m.f.) of 𝑿
𝒇(𝒙) 𝟏Τ𝟒 𝟐Τ𝟒 𝟏Τ𝟒 1
Bivariate (Joint)
Distributions
(discrete)
 Bivariate (Joint) Distributions
• If 𝑿 and 𝒀 is a discrete r.v.’s with possible pairs of values

𝒙𝒊 , 𝒚𝒋 , 𝒊 = 𝟏, 𝟐, . . . . ; 𝒋 = 𝟏, 𝟐, … , then
𝒉 𝒙𝒊 , 𝒚𝒋 = 𝑷 𝑿 = 𝒙𝒊 , 𝒀 = 𝒚𝒋 , 𝒙 = 𝒙𝟏 , 𝒙𝟐 , … ; 𝒚 = 𝒚𝟏 , 𝒚𝟐 , …
is called the joint probability mass function of 𝑿 and 𝒀.

• 𝑷 𝑿 = 𝒙𝒊 , 𝒀 = 𝒚𝒋 is the probability of the intersection of

the events 𝑿 = 𝒙𝒊 and 𝒀 = 𝒚𝒋 (i.e., together or at the

same time).
Bivariate (Joint) Distributions

𝑿 𝒙𝟏 𝒙𝟐 𝒙𝟑 ...
𝒀
𝒚𝟏 𝒉(𝒙𝟏 , 𝒚𝟏 ) 𝒉(𝒙𝟐 , 𝒚𝟏 ) 𝒉(𝒙𝟑 , 𝒚𝟏 ) ...

𝒚𝟐 𝒉(𝒙𝟏 , 𝒚𝟐 ) 𝒉(𝒙𝟐 , 𝒚𝟐 ) 𝒉(𝒙𝟑 , 𝒚𝟐 ) ...

𝒚𝟑 𝒉(𝒙𝟏 , 𝒚𝟑 ) 𝒉(𝒙𝟐 , 𝒚𝟑 ) 𝒉(𝒙𝟑 , 𝒚𝟑 ) ...

... ... ... ... ...

 Bivariate (Joint) Distributions

• A bivariate function 𝒉(𝒙, 𝒚) can serve as a joint p.m.f. of a pair of r.v.'s 𝑿

and 𝒀 iff it satisfies the conditions

1. 𝒉 𝒙, 𝒚 ≥ 𝟎 for all 𝒙 and 𝒚

2. σ𝒊 σ𝒋 𝒉 𝒙𝒊 , 𝒚𝒋 = 𝟏
 Bivariate (Joint) Distributions
𝑿 𝒙𝟏 𝒙𝟐 𝒙𝟑 ... Sum
𝒀
𝒚𝟏 𝒉(𝒙𝟏 , 𝒚𝟏 ) 𝒉(𝒙𝟐 , 𝒚𝟏 ) 𝒉(𝒙𝟑 , 𝒚𝟏 ) ... The conditions

𝒚𝟐 𝒉(𝒙𝟏 , 𝒚𝟐 ) 𝒉(𝒙𝟐 , 𝒚𝟐 ) 𝒉(𝒙𝟑 , 𝒚𝟐 ) ...

1. 𝒉 𝒙, 𝒚 ≥ 𝟎
𝒚𝟑 𝒉(𝒙𝟏 , 𝒚𝟑 ) 𝒉(𝒙𝟐 , 𝒚𝟑 ) 𝒉(𝒙𝟑 , 𝒚𝟑 ) ...
2. σ𝒊 σ𝒋 𝒉 𝒙𝒊 , 𝒚𝒋 = 𝟏
... ... ... ... ...

Sum 1
 Example 1.1

Determine the value of the constant 𝒌 so that the following

function,
𝒉(𝒙, 𝒚) = 𝒌 𝒙 𝒚 𝒇𝒐𝒓 𝒙 = 𝟏, 𝟐, 𝟑 ; 𝒚 = 𝟏, 𝟐, 𝟑

can serve as a joint p.m.f. of two r.v.'s 𝑿 and 𝒀, then find

𝑷 𝑿 + 𝒀≥𝟒 .
 Example 1.1
𝒉(𝒙, 𝒚) = 𝒌 𝒙 𝒚 𝒇𝒐𝒓 𝒙 = 𝟏, 𝟐, 𝟑 ; 𝒚 = 𝟏, 𝟐, 𝟑

• From the first condition, the constant k must be nonnegative.

• To satisfy the second condition, we must have
෍ ෍ 𝒉 𝒙𝒊 , 𝒚𝒋 = 𝟏
𝒊 𝒋
 Example 1.1
𝒉(𝒙, 𝒚) = 𝒌 𝒙 𝒚 𝒇𝒐𝒓 𝒙 = 𝟏, 𝟐, 𝟑 ; 𝒚 = 𝟏, 𝟐, 𝟑

𝑿 𝟏 𝟐 𝟑 Sum
𝒀
𝟏 𝒌 𝟐𝒌 𝟑𝒌 𝟔𝒌

𝟐 𝟐𝒌 𝟒𝒌 𝟔𝒌 𝟏𝟐 𝒌

𝟑 𝟑𝒌 𝟔𝒌 𝟗𝒌 𝟏𝟖 𝒌

Sum 𝟑𝟔 𝒌
 Example 1.1
𝒉(𝒙, 𝒚) = 𝒌 𝒙 𝒚 𝒇𝒐𝒓 𝒙 = 𝟏, 𝟐, 𝟑 ; 𝒚 = 𝟏, 𝟐, 𝟑

• From the first condition, the constant k must be nonnegative.

• To satisfy the second condition, we must have
෍ ෍ 𝒉 𝒙𝒊 , 𝒚𝒋 = 𝟏.
𝒊 𝒋
Thus, we must have
𝟑𝟔 𝒌 = 𝟏
and therefore
𝟏
𝒌 = .
𝟑𝟔
 Example 1.1

𝑿 𝟏 𝟐 𝟑
𝒀
𝟏 𝟐 𝟑
𝟏
𝟑𝟔 𝟑𝟔 𝟑𝟔
𝟐 𝟒 𝟔
𝟐
𝟑𝟔 𝟑𝟔 𝟑𝟔
𝟑 𝟔 𝟗
𝟑
𝟑𝟔 𝟑𝟔 𝟑𝟔
 Example 1.1

Determine the value of the constant 𝒌 so that the following

function,
𝒉(𝒙, 𝒚) = 𝒌 𝒙 𝒚 𝒇𝒐𝒓 𝒙 = 𝟏, 𝟐, 𝟑 ; 𝒚 = 𝟏, 𝟐, 𝟑

can serve as a joint p.m.f. of two r.v.'s 𝑿 and 𝒀, then find

𝑷 𝑿 + 𝒀≥𝟒 .
 Example 1.1
𝑷 𝑿 + 𝒀≥𝟒

𝑿 𝟏 𝟐 𝟑
𝒀
𝟏 𝟐 𝟑 𝑷 𝑿 + 𝒀≥𝟒 =
𝟏
𝟑𝟔 𝟑𝟔 𝟑𝟔 𝒉 𝟏, 𝟑 + 𝒉 𝟐, 𝟐 + 𝒉 𝟐, 𝟑 +
𝟐 𝟒 𝟔 𝒉 𝟑, 𝟏 + 𝒉 𝟑, 𝟐 + 𝒉 𝟑, 𝟑 =
𝟐
𝟑𝟔 𝟑𝟔 𝟑𝟔 𝟑𝟏/𝟑𝟔
𝟑 𝟔 𝟗
𝟑
𝟑𝟔 𝟑𝟔 𝟑𝟔
 Example 1.1
𝑷 𝑿 + 𝒀≥𝟒

𝑿 𝟏 𝟐 𝟑
𝒀
𝑷 𝑿 + 𝒀≥𝟒 =
𝟏 𝟐 𝟑
𝟏 𝟏 − 𝑷 𝑿 + 𝒀 < 𝟒 =
𝟑𝟔 𝟑𝟔 𝟑𝟔
𝟏 − 𝒉 𝟏, 𝟏 + 𝒉 𝟏, 𝟐 + 𝒉 𝟐, 𝟏
𝟐 𝟒 𝟔
𝟐 = 𝟏 − {𝟏/𝟑𝟔 + 𝟐/𝟑𝟔 + 𝟐/𝟑𝟔}
𝟑𝟔 𝟑𝟔 𝟑𝟔
= 𝟑𝟏/𝟑𝟔.
𝟑 𝟔 𝟗
𝟑
𝟑𝟔 𝟑𝟔 𝟑𝟔
 Marginal Distributions
• If 𝑿 and 𝒀 is a discrete r.v.’s with joint p.m.f. 𝒉(𝒙, 𝒚), the
p.m.f. , the marginal p.m.f. of X is given by

𝒈𝑿 𝒙𝒊 = ෍ 𝒉 𝒙𝒊 , 𝒚𝒋 , 𝒊 = 𝟏, 𝟐, … ;
𝒋

• Correspondingly, the following function

𝒈𝒀 𝒚𝒋 = ෍ 𝒉 𝒙𝒊 , 𝒚𝒋 , 𝒋 = 𝟏, 𝟐, … ;
𝒊

is called the marginal p.m.f. of Y.

 Example 1.1
Using the previous example (Example1.1), find the probability
distributions of 𝑿 alone (marginal pmf of 𝑿) and that of 𝒀
alone(marginal pmf of 𝒀).
𝑿 𝟏 𝟐 𝟑
𝒀
𝟏 𝟐 𝟑
𝟏
𝟑𝟔 𝟑𝟔 𝟑𝟔
𝟐 𝟒 𝟔
𝟐
𝟑𝟔 𝟑𝟔 𝟑𝟔
𝟑 𝟔 𝟗
𝟑
𝟑𝟔 𝟑𝟔 𝟑𝟔
 Example 1.1
𝑿 𝟏 𝟐 𝟑 Sum
𝒀
𝟏 𝟐 𝟑 𝟔
𝟏
𝟑𝟔 𝟑𝟔 𝟑𝟔 𝟑𝟔
𝟐 𝟒 𝟔 𝟏𝟐
𝟐
𝟑𝟔 𝟑𝟔 𝟑𝟔 𝟑𝟔
𝟑 𝟔 𝟗 𝟏𝟖
𝟑
𝟑𝟔 𝟑𝟔 𝟑𝟔 𝟑𝟔
𝟔 𝟏𝟐 𝟏𝟖
Sum 𝟏
𝟑𝟔 𝟑𝟔 𝟑𝟔
 Example 1.1
The marginal pmf of 𝑿

𝑿 𝟏 𝟐 𝟑 Sum
𝟔 𝟏𝟐 𝟏𝟖
Sum 𝟏
𝟑𝟔 𝟑𝟔 𝟑𝟔
The marginal pmf of 𝒀

𝒀 𝟏 𝟐 𝟑 Sum
𝟔 𝟏𝟐 𝟏𝟖
Sum 𝟏
𝟑𝟔 𝟑𝟔 𝟑𝟔
 Example
Two textbooks are selected at random from a shelf contains 3
statistics texts, 2 mathematics texts and 3 arts texts. If 𝑿 is the
number of statistics texts and 𝒀 the number of mathematics texts
actually chosen.

a) Find the joint distribution of 𝑿 and 𝒀.

b) Find the probability distributions of 𝑿 alone (marginal pmf of 𝑿)

and that of 𝒀 alone(marginal pmf of 𝒀).
 Example
3 statistics texts, 2 mathematics texts and 3 arts texts

𝑿 𝒀

• But 𝑿+𝒀≤𝟐
• Then, all possible pairs of 𝑋 and 𝑌 are:
(0, 0) , (0,1) , (1,0) , (1,1) , (0,2) 𝑎𝑛𝑑 (2,0)
 Example

For example, to find the probability associated with (𝟏, 𝟎),

1 of the 3 statistics texts
i.e., the event of choosing 0 of the 2 math. texts

1 of the 3 arts texts

The number of ways in which this can be done is

𝟑 𝟐 𝟑
=𝟗
𝟏 𝟎 𝟏
 Example

• The total number of ways in which 2 of the 8 texts can be

𝟖
chosen is = 𝟐𝟖.
𝟐
• Since, the textbooks are selected at random (equally
likely), the probability associated with (𝟏, 𝟎) is

𝟗
𝒉 𝟏, 𝟎 = 𝑷 𝑿 = 𝟏 , 𝒀 = 𝟎 = .
𝟐𝟖
 Example

• Similarly, the probability associated with (𝟏, 𝟏) is

𝟑 𝟐 𝟑
𝒉 𝟏, 𝟏 = 𝑷 𝑿 = 𝟏 , 𝒀 = 𝟏 = 𝟏 𝟏 𝟎 = 𝟑 .
𝟖 𝟏𝟒
𝟐
 Example
Then, the joint p.m.f. of the r.v.'s 𝑿 and 𝒀 is given by
the following table
𝑿
𝟎 𝟏 𝟐
𝒀

𝟎 𝟑/𝟐𝟖 𝟗/𝟐𝟖 𝟑/𝟐𝟖

𝟏 𝟑/𝟏𝟒 𝟑/𝟏𝟒 𝟎

𝟐 𝟏/𝟐𝟖 𝟎 𝟎
 Example

• Actually, it is generally preferable to represent

probabilities such as these by means of a formula.

𝟑 𝟐 𝟑
𝒙 𝒚
𝟐−𝒙−𝒚
𝒉 𝒙, 𝒚 = 𝑷 𝑿 = 𝒙 , 𝒀 = 𝒚 = ,
𝟖
𝟐
𝒙 = 𝟎, 𝟏, 𝟐 ; 𝒚 = 𝟎, 𝟏, 𝟐; 𝒙 + 𝒚 ≤ 𝟐.
 Example
To find the marginal pmf
𝑿 𝟎 𝟏 𝟐 Sum
𝒀
𝟏𝟓
𝟎 𝟑/𝟐𝟖 𝟗/𝟐𝟖 𝟑/𝟐𝟖
𝟐𝟖
𝟏𝟐
𝟏 𝟑/𝟏𝟒 𝟑/𝟏𝟒 𝟎
𝟐𝟖
𝟏
𝟐 𝟏/𝟐𝟖 𝟎 𝟎
𝟐𝟖
𝟏𝟎 𝟏𝟓 𝟑
Sum 𝟏
𝟐𝟖 𝟐𝟖 𝟐𝟖
 Example
The marginal pmf of 𝑿

𝑿 𝟎 𝟏 𝟐 Sum
𝟏𝟎 𝟏𝟓 𝟑
Sum 𝟏
𝟐𝟖 𝟐𝟖 𝟐𝟖
The marginal pmf of 𝒀

𝑿 𝟎 𝟏 𝟐 Sum
𝟏𝟓 𝟏𝟐 𝟏
Sum 𝟏
𝟐𝟖 𝟐𝟖 𝟐𝟖