Random Variables

A random variable (RV) X is a function from sample space S to set of real

numbers.
It can be classified into two,(i) discrete RV, (ii) continuous RV

Discrete random variable:

The RV, X is said to be discrete if X has finite values or countably infinite
values

Probability distribution/ Probability Mass function/Probability

function (pmf)
Let 𝑋 = 𝑥1 , 𝑥2 , 𝑥3 , 𝑥4 , … be a discrete RV. The function P is said to be pmf if
(i) 𝑃(𝑥 = 𝑥𝑖 ) = 𝑝𝑖 ≥ 0, 𝑖 = 1,2,3, …
(ii) ∑𝑖 𝑝𝑖 = 1
That is the pmf is
𝑥𝑖 𝑥1 𝑥2 𝑥3 𝑥4 𝑥5 𝑥6 …. 𝑥𝑛−1 𝑥𝑛
𝑝𝑖 𝑝1 𝑝2 𝑝3 𝑝4 𝑝5 𝑝6 …. 𝑝𝑛−1 𝑝𝑛

Cumulative distribution function (cdf)

Let 𝑋 = 𝑥1 , 𝑥2 , 𝑥3 , 𝑥4 , … be a discrete RV. The cdf of X is defined as
𝐹(𝑥) = 𝑃(𝑋 ≤ 𝑥) = 𝑃(−∞ < 𝑋 ≤ 𝑥)

𝐹(𝑥) = ∑ 𝑃(𝑋 = 𝑥𝑖 )
𝑥𝑖 ≤𝑥

For example if, 𝑥 ≤ 𝑥5 , then

𝐹(𝑥) = ∑ 𝑃(𝑋 = 𝑥𝑖 ) = 𝑝1 + 𝑝2 + 𝑝3 + 𝑝4 + 𝑝5
𝑥𝑖 ≤𝑥

if, 𝑥 < 𝑥5 , then

 𝐹(𝑥) = ∑ 𝑃(𝑋 = 𝑥𝑖 ) = 𝑝1 + 𝑝2 + 𝑝3 + 𝑝4
𝑥𝑖 ≤𝑥

Continuous random variable:

The RV, X is said to be continuous if X is defined on an
interval like: (a,b), (-∞,a), (a,∞) or (-∞,∞)

Probability distribution/ Probability density

function/Probability function (pdf)
Let 𝑋 be a continuous RV. The function f is said to be pdf if
(i) 𝑓(𝑥) ≥ 0,
∞
(ii) ∫−∞ 𝑓(𝑥)𝑑𝑥 = 1

Cumulative distribution function (cdf)

Let 𝑋 be a continuous RV. The cdf of X is defined as
𝐹(𝑥) = 𝑃(𝑋 ≤ 𝑥) = 𝑃(−∞ < 𝑋 ≤ 𝑥)
𝑥
𝐹(𝑥) = 𝑃(−∞ < 𝑋 ≤ 𝑥) = ∫ 𝑓(𝑡)𝑑𝑡
−∞

Properties of cdf
1. F(x) is non decreasing function , that is 𝑥𝑖 < 𝑥𝑗 , 𝑡ℎ𝑒𝑛 𝐹(𝑥𝑖 ) ≤ 𝐹(𝑥𝑗 )
2. 𝐹(−∞) = 0; 𝐹(∞) = 1
3. If X is discrete RV, then 𝑃(𝑋 = 𝑥𝑖 ) = 𝐹(𝑥𝑖 ) − 𝐹(𝑥𝑖−1 )
4. If X is continuous RV then
𝑏 𝑎
𝑃(𝑎 < 𝑋 < 𝑏) = 𝐹(𝑏) − 𝐹(𝑎) = ∫ 𝑓(𝑡)𝑑𝑡 − ∫ 𝑓(𝑡)𝑑𝑡 ;
−∞ −∞

𝑃(𝑎 ≤ 𝑋 ≤ 𝑏) = 𝐹(𝑏) − 𝐹(𝑎),

𝑃(𝑎 ≤ 𝑋 < 𝑏) = 𝐹(𝑏) − 𝐹(𝑎);
𝑃(𝑎 < 𝑋 ≤ 𝑏) = 𝐹(𝑏) − 𝐹(𝑎)
5. If X is continuous the 𝑓(𝑥) = 𝑑𝐹
𝑑𝑥

Expected Values
Let X be a random variable (R.V). The expected value of X is defined as
⮚ If X is discrete RV then 𝐸(𝑋) = ∑𝑖 𝑥𝑖 𝑃(𝑋 = 𝑥𝑖 )
∞
⮚ If X is discrete RV then 𝐸(𝑋) = ∫−∞ 𝑥𝑓(𝑥)𝑑𝑥
Properties
1. 𝐸(𝑎) = 𝑎
2. 𝐸(𝑎𝑋 + 𝑏) = 𝑎𝐸(𝑋) + 𝑏
2
3. 𝜎2 = 𝑉𝑎𝑟(𝑋) = 𝐸(𝑋 2 ) − (𝐸 (𝑋))
4. 𝑉𝑎𝑟(𝑎𝑋 + 𝑏) = 𝑎2 𝑉𝑎𝑟(𝑋)

Moments
A. The rth moment about origin is denoted as 𝜇′𝑟 = 𝐸(𝑋 𝑟 )
⮚ If X is discrete RV then 𝜇′𝑟 = 𝐸(𝑋 𝑟 ) = ∑𝑖 𝑥𝑖𝑟 𝑃(𝑋 = 𝑥𝑖 )
∞
⮚ If X is continuous RV then 𝜇′𝑟 = 𝐸(𝑋 𝑟 ) = ∫−∞ 𝑥 𝑟 𝑓(𝑥)𝑑𝑥
B. The rth moment about mean is denoted as 𝜇𝑟 = 𝐸[(𝑋 − 𝑥̅ )𝑟 ]
⮚ If X is discrete RV then 𝜇𝑟 = 𝐸[(𝑋 − 𝑥̅ )𝑟 ] = ∑𝑖 (𝑥𝑖 − 𝑥̅ )𝑟 𝑃(𝑋 = 𝑥𝑖 )
∞
⮚ If X is continuous RV then 𝜇𝑟 = 𝐸[(𝑋 − 𝑥̅ )𝑟 ] = ∫−∞(𝑥 − 𝑥̅ )𝑟 𝑓(𝑥)𝑑𝑥

𝜇1 = 𝐸(𝑋 − 𝑥̅ ) = 𝐸(𝑋) − 𝑥̅ = 𝑥̅ − 𝑥̅ = 0
2
𝜇2 = 𝐸[(𝑋 − 𝑥̅ )2 ] = 𝐸(𝑋 2 ) − 𝐸(𝑋)2 = 𝜇2′ − 𝜇1′
𝜇3 = 𝐸[(𝑋 − 𝑥̅ )3 ] = 𝐸(𝑋 3 ) − 3𝐸(𝑋 2 )𝑥̅ + 3𝐸(𝑋)𝑥̅ 2 − 𝑥̅ 3
3
= 𝜇3′ − 3𝜇2′ 𝜇1′ + 2𝜇1′
𝜇4 = 𝐸[(𝑋 − 𝑥̅ )4 ] = 𝐸(𝑋 4 ) − 4𝐸(𝑋 3 )𝑥̅ + 6𝐸(𝑋 2 )𝑥̅ 2 − 4𝑥̅ 3 𝐸(𝑋) + 𝑥̅ 4
2 4
= 𝜇4′ − 4𝜇3′ 𝜇1′ + 6𝜇1′ 𝜇2′ − 3𝜇1′

C. The rth moment about a point A is denoted as 𝑚𝑟 = 𝐸[(𝑋 − 𝐴)𝑟 ]

𝑚1 = 𝐸(𝑋 − 𝐴) = 𝑥̅ − 𝐴
𝑚2 = 𝐸[(𝑋 − 𝐴)2 ] = 𝐸(𝑋 2 ) − 2𝐴𝐸(𝑋) + 𝐴2
𝑚3 = 𝐸[(𝑋 − 𝐴)3 ] = 𝐸(𝑋 3 ) − 3𝐸(𝑋 2 )𝐴 + 3𝐸(𝑋)𝐴2 − 𝐴3
𝑚4 = 𝐸[(𝑋 − 𝐴)4 ] = 𝐸(𝑋 4 ) − 4𝐸(𝑋 3 )𝐴 + 6𝐸(𝑋 2 )𝐴2 − 4𝐴3 𝐸(𝑋) + 𝐴4
 Moment generating function MGF
𝑡𝐸(𝑋) 𝑡 2 𝐸(𝑋 2 ) 𝑡 3 𝐸(𝑋 3 )
The MGF of discrete RV is 𝑀𝑋 (𝑡) = 𝐸(𝑒 𝑡𝑋 ) = 1 + + + +
1! 2! 3!
𝑡 4 𝐸(𝑋 4 )
+⋯
4!

𝑡𝜇1′ 𝑡 2 𝜇2′ 𝑡 3 𝜇3′ 𝑡 4 𝜇4′

=1+ + + + +⋯
1! 2! 3! 4!
∞
𝑡 𝑟 𝜇𝑟′
=∑
𝑟!
𝑟=0

Properties :

𝑡𝑟
I. 𝜇′𝑟 𝑟
= 𝐸(𝑋 ) is the coefficient of
𝑟!
𝑑𝑟
II. 𝜇′𝑟 = 𝐸(𝑋 𝑟 ) = [ 𝑀𝑋 (𝑡)]
𝑑𝑡 𝑟 𝑡=0
III. If the MGF of X is 𝑀𝑋 (𝑡) and if 𝑌 = 𝑎𝑋 + 𝑏 then 𝑀𝑌 (𝑡) = 𝑒 𝑏𝑡 𝑀𝑋 (𝑎𝑡)

Problems:

1. Two fair coins are tossed simultaneously. The RV X is the number of heads
appear. The find the following: (i) pmf, (ii) mean and variance, (iii) mgf,
(iv) E(2X+3), (v) cdf
Solution:
The sample space S={TT,TH,HT,HH}
X= the number of heads appears= 0,1,2
Therefore (i) the probability mass function is

x 0 1 2

p 1/4 2/4 1/4

(ii) For mean and variance

x x2 p xp x2p

0 0 1/4 0 0
 1 1 1/2 1/2 1/2

2 4 1/4 1/2 1

Summation 1 1 3/2

Therefore the mean 𝜇1′ = 𝑥̅ = ∑ 𝑥𝑝 = 1

2
The Variance 𝜎2 = 𝑉𝑎𝑟(𝑋) = 𝐸(𝑋 2 ) − (𝐸 (𝑋))
2 3 1
𝜎 2 = ∑ 𝑥 2 𝑝 − (∑ 𝑥𝑝) = − 1 =
2 2

(iii) The mgf 𝑀𝑋 (𝑡) = 𝐸(𝑒 𝑡𝑋 )

2
1 1 1
𝑀𝑋 (𝑡) = 𝐸(𝑒 𝑡𝑋 ) = ∑ 𝑒 𝑡𝑘 𝑃(𝑥 = 𝑘) = 𝑒 0∗𝑡 + 𝑒 1∗𝑡 + 𝑒 2∗𝑡
4 2 4
𝑘=0

1 1 1
= + 𝑒 𝑡 + 𝑒 2𝑡
4 2 4

(iv) E(2X+3)=2E(X)+3=2*1+3=5

(v) 𝐹(𝑥) = 0, 𝑥<0

1
= , 𝑥<1
4
1 2
= + , 𝑥<2
4 4
1 2 1
= + + ,𝑥 > 2
4 4 4
2. A rv X has the following probability function
𝑥
0 1 2 3 4 5 6 7
𝑝 0 𝑘 2𝑘 2𝑘 3𝑘 𝑘2 2𝑘 2 7𝑘 2 + 𝑘
 Find the following: (i) The value of k (ii) 𝑃(𝑋 < 6); 𝑃(𝑋 ≥ 6), (iii) the
1
smallest value of 𝛼 for which 𝑃(𝑋 ≤ 𝛼) > (iv) 𝑃(1.5 < 𝑋 < 4.5/𝑋 > 2)
2
(v) The mean and variance
Solution:
(i) We know that ∑𝑖 𝑝𝑖 = 1
Therefore ∑𝑖 𝑝𝑖 = 𝑘 + 2𝑘 + 2𝑘 + 3𝑘 + 𝑘 2 + 2𝑘 2 + 7𝑘 2 + 𝑘 = 1
10𝑘 2 + 9𝑘 − 1 = 0 ⇒ 𝑘 = 1/10

(ii) 𝑃(𝑋 < 6) = 𝑃(𝑋 = 0) + 𝑃(𝑋 = 1) + 𝑃(𝑋 = 2) + 𝑃(𝑋 = 3) +

81
+𝑃(𝑋 = 4) + 𝑃(𝑋 = 5) = 𝑘 2 + 8𝑘 =
100
19
𝑃(𝑋 ≥ 6) = 𝑃(𝑋 = 6) + 𝑃(𝑋 = 7) = 9𝑘 2 + 𝑘 =
100
4.5 𝑃([1.5<𝑋<4.5]∩[𝑋>2]) 𝑃(𝑋=3)+𝑃(𝑋=4)
(iii) 𝑃 (1.5 < 𝑋 < > 2) = = =
𝑋 𝑃(𝑋>2) 𝑃(𝑋=3,4,5,6,7)

(iv) The mean 𝑥̅ = ∑𝑖 𝑥𝑖 𝑝𝑖 and the variance 𝜎 2 = 𝜇2′ − (𝜇1 )2

x X2 p xp X2p
0 0 0 0 0
1 1 1/10 1/10 10/100
2 4 2/10 4/10 80/100
3 9 2/10 6/10 180/100
4 16 3/10 12/10 480/100
5 25 1/100 5/100 25/100
6 36 2/100 12/100 72/100
7 49 17/100 119/100 833/100
Summation 366/100 1680/100
366 1680 366 2
𝑥̅ = ∑𝑖 𝑥𝑖 𝑝𝑖 = ; 𝜎2 = −( )
100 100 100

3. The first four moments about the origin are 2, 6, 22, 124 then find the first
four moments about the mean.
Solution
𝜇1′ = 2; 𝜇2′ = 6; 𝜇3′ = 22; 𝜇4′ = 124 then

𝜇1 = 0
𝜇2 = 𝜇2′ − (𝜇1′ )2 = 6 − 22 = 2
𝜇3 = 𝜇3′ − 3𝜇2′ 𝜇1′ + 2(𝜇1′ )3 = 22 − 3 × 6 × 2 + 2 × 23 = 2
2
𝜇4 = 𝜇4′ − 4𝜇3′ 𝜇1′ + 6𝜇2′ 𝜇1′ − 3(𝜇1′ )3
= 124 − 4 × 22 × 2 + 6 × 6 × 4 − 3 × 16 = 44
 𝑘𝑒 −𝑥 , 𝑥 ≥ 0
4. The pdf of X is 𝑓(𝑥) = { . Then find the value of k, first 3
0, 𝑥 < 0
moments about origin and hence first 3 moments about mean.
Solution
∞
∫ 𝑓(𝑥)𝑑𝑥 = 1
−∞

∞ ∞
= ∫ 𝑓(𝑥)𝑑𝑥 = ∫ 𝑘𝑒 −𝑥 𝑑𝑥 = − 𝑘[𝑒 −𝑥 ]∞
0 =𝑘 =1
−∞ 0

∞
𝜇1′ = 𝐸(𝑋) = ∫ 𝑥𝑒 −𝑥 𝑑𝑥 = [𝑥(−𝑒 −𝑥 ) − 1(𝑒 −𝑥 )]∞
0 =1
0
∞
𝜇2′ = 𝐸(𝑋 2)
= ∫ 𝑥 2 𝑒 −𝑥 𝑑𝑥 = [𝑥 2 (−𝑒 −𝑥 ) − (2𝑥)(𝑒 −𝑥 )+2(−𝑒 −𝑥 )]∞
0 =2
0
∞
𝜇3′ = 𝐸(𝑋 3 ) = ∫ 𝑥 3 𝑒 −𝑥 𝑑𝑥 = [𝑥 3 (−𝑒 −𝑥 )
0
− (3𝑥 2 )(𝑒 −𝑥 )+6𝑥(−𝑒 −𝑥 ) − 6(𝑒 −𝑥 )]∞
0 =6
Hence
𝜇1 = 0
𝜇2 = 𝜇2′ − (𝜇1′ )2 = 2 − 12 = 1
𝜇3 = 𝜇3′ − 3𝜇2′ 𝜇1′ + 2(𝜇1′ )3 = 6 − 3 × 2 × 1 + 2 × 13 = 2
5. The pdf of X is 𝑓(𝑥) = 𝑘𝑒 −|𝑥| , −∞ < 𝑥 < ∞. Then find the value of k,
first 3 moments about origin and hence first 3 moments about mean.