TOPICS TO BE COVERED

• BINOMIAL DISTRIBUTION
• POISSON DISTRIBUTION
• NORMAL DISTRIBUTION
• GAMMA DISTRIBUTION
• EXPONENTIAL DISTRIBUTION
• WEIBULL DISTRIBUTION
 Special Distribution
Discrete Distribution:
• Binomial
• Poisson
Continuous Distribution:
• Normal
• Gamma
• Exponential
• Weibull
 Bernoulli Distribution
• Any experiments whose outcome is 0, 1 can be modelled as Bernoulli Distribution,
• 𝑋~𝐵𝑒𝑟𝑛𝑜𝑢𝑙𝑙𝑖 𝑝 if its PMF 𝑝 𝑥 = 1 = 𝑝, 𝑝 𝑥 = 0 = 1 − 𝑝, 0 < 𝑝 < 1
• Mean: 𝐸 𝑋 = 𝑥𝑝 𝑥 = 1 𝑝 + 0 1 − 𝑝 = 𝑝
• 𝐸 𝑋2 = 𝑥 2𝑝 𝑥 = 𝑝
2
• Variance, 𝑣𝑎𝑟 𝑥 = 𝐸 𝑋 2 − 𝐸 𝑋 = 𝑝 − 𝑝2 = 𝑝(1 − 𝑝)
𝑥𝑖 𝑡 𝑝(𝑥 )
• MGF, 𝑀𝑋 𝑡 = 𝑖𝑒 𝑖 = 𝑒 0𝑡 1 − 𝑝 + 𝑒 𝑡 𝑝 = 𝑒 𝑡 𝑝 + (1 − 𝑝)
 Binomial Distribution
• 𝑃 𝑋 = 𝑖 = 𝑛𝐶𝑖 𝑝𝑖 𝑞 𝑛;𝑖
• Mean, 𝐸 𝑥 = 𝑛𝑝
• 𝐸 𝑋 2 = 𝑛 𝑛 − 1 𝑝2 + 𝑛𝑝
• 𝑣𝑎𝑟 𝑥 = 𝑛𝑝𝑞
• 𝑀𝑋 𝑡 = 𝑝𝑒 𝑡 + 𝑞 𝑛
 Problem
1. Compute the mean & variance of the number of heads in 3 flips of a fair coin.
Solution:
𝑋~Binomial Distribution
1 1
𝑛 = 3, 𝑝 = 2 , 𝑞 = 2
1 3
Mean = 𝑛𝑝 = 3 =2
2
1 1 3
Variance = 𝑛𝑝𝑞 = 3 =4
2 2
2. Let 𝑋 be a binomial random variable with mean 7 and variance 2.1. Find 𝑃(𝑋 = 4).
Solution:
Given 𝑛𝑝 = 7 ----- (1)
𝑛𝑝𝑞 = 2.1 ----- (2)
Sub.(1) in (2) ⇒ 7. 𝑞 = 2.1 ⇒ 𝑞 = 0.3
⇒ 𝑝 = 0.7
(1)⇒ 𝑛 × 0.7 = 7
𝑛 = 10
∴ 𝑃 𝑋 = 4 = 10𝐶4 0.7 4 0.3 6 = 0.03676
 𝑝 𝑛;𝑘
3. If 𝑋~Binomial 𝑛, 𝑝 . 𝑆. 𝑇 𝑃 𝑋 = 𝑘 + 1 = 𝑃(𝑋 = 𝑘), 𝑘 = 0,1, … , 𝑛
1;𝑝 𝑘:1

Proof
𝑛
𝑃 𝑋 =𝑘+1 = 𝑝𝑘:1 𝑞 𝑛;𝑘;1 ; 𝑘 = 0, 1, … , 𝑛 − 1
𝑘+1
 Poisson Distribution
• 𝑋 – R.V. in which no. of occurrence of a particular event can be modelled as Poisson Distribution
𝑒 −𝜆 𝜆𝑖
• 𝑃 𝑋=𝑖 = 𝑖!
,𝑖 = 0, 1, …

• If 𝑛 is very large in Binomial Distribution is tedious. Hence for it Poisson distribution is indulged
Poisson Approx. to a Binomial Distribution:
If 𝑛 is large and 𝑝 is small, consider 𝜆 = 𝑛𝑝
−𝜆 𝑖
𝑒 𝜆 𝑖−1
∞ ∞ 𝜆
• Mean, 𝐸 𝑋 = 𝑖 𝑖. 𝑃 𝑋=𝑖 = 𝑖<0 𝑖. = 𝑒 ;𝜆 . 𝜆 𝑖<0 𝑖;1 ! = 𝑒 ;𝜆 . 𝜆. 𝑒 𝜆 = 𝜆
𝑖!

• Mean, 𝐸 𝑋 = 𝜆
 2 −𝜆 𝑖 2 −𝜆 𝑖
∞ 𝑖 𝑒 𝜆 ∞ (𝑖 ;𝑖:𝑖) 𝑒 𝜆
• 𝐸 𝑋2 = 𝑖<0 = 𝑖<0
𝑖! 𝑖!

𝜆𝑖 𝑖
⇒𝐸 𝑋2 = ∞
𝑒 ;𝜆 . 𝑖! ∞ ;𝜆 . 𝜆
𝑖<0 𝑖 𝑖−1 + 𝑖<0 𝑖. 𝑒 𝑖!

⇒ 𝐸 𝑋 2 = 𝑒 ;𝜆 𝜆2 𝑒 𝜆 + 𝜆 = 𝜆2 + 𝜆
⇒ 𝐸 𝑋 2 = 𝜆2 + 𝜆
2
• Variance, 𝑉𝑎𝑟 𝑋 = 𝐸 𝑋 2 − 𝐸 𝑋 =𝜆
−𝜆 𝑖
∞ 𝑖𝑡 . 𝑒 𝜆 𝑡𝜆 1;𝑒 𝑡
• MGF, 𝑀𝑋 𝑡 = 𝑖<0 𝑒 = 𝑒 ;𝜆 . 𝑒 𝑒 = 𝑒 ;𝜆
𝑖!

Remark:
It is only distribution in Discrete whose mean and variance are same
 Problem
1. If 𝑋 is a Poisson random variable such that 𝑃 𝑋 = 2 = 9 𝑃 𝑋 = 4 + 90𝑃(𝑋 = 6). Calculate the mean
and variance of 𝑋.
Solution:
(𝑒 −𝜆 𝜆𝑖 )
𝑃 𝑋=𝑖 = 𝑖!
, i=0, 1, 2, …
𝑃 𝑋=2 =9 𝑃 𝑋=4 + 90𝑃(𝑋 = 6)
(𝑒 −𝜆 𝜆2 ) (𝑒 −𝜆 𝜆4 ) (𝑒 −𝜆 𝜆6 )
⇒ 2!
=9 4!
= 90 6!
𝜆2 9𝜆4 90𝜆6
⇒ 2
= 24
= 720
3𝜆2 𝜆4
⇒1= 8
+ 4
⇒ 𝜆4 + 3𝜆2 − 4 = 0
⇒ 𝜆2 + 4 𝜆2 − 1 = 0
⇒𝜆=1
𝐸 𝑋 = 𝑉𝑎𝑟 𝑋 = 1
2. Suppose 𝑋 is a random variable with mean & variance both equal to 6. What is 𝑃 𝑋 < 3 ?
Solution
Given 𝑀𝑒𝑎𝑛 = 𝑣𝑎𝑟𝑖𝑎𝑛𝑐𝑒 = 6
𝑋~Poisson (𝜆 = 6)
𝑃 𝑋 < 3 = 𝑃 𝑋 = 0 + 𝑃 𝑋 = 1 + 𝑃(𝑋 = 2)
0
6 1
6 62
⇒𝑃 𝑋<3 = 𝑒 ;6 . 0! + 𝑒 ;6 . 1! + 𝑒 ;6 . 2!

⇒ 𝑃 𝑋 < 3 = 𝑒 ;6 1 + 6 + 18 = 0.06197
3. The probability that a person dies due to viral infection is 0.001 of the next 4000 person infected. What
is the mean no. who will die?
Solution:
Here 𝑋~ Poisson Distribution
𝑛 = 4000, 𝑝 = 0.001
Poisson Approx. to Binomial 𝜆 = 𝑛𝑝 = 4
Mean no. of person will die is 4
 Continuous Distribution
• Exponential Distribution
• Normal Distribution
• Gamma Distribution
• Weibull Distribution
 Exponential Distribution
• 𝑋~ exp(𝜆) , 𝜆 ≥ 0, P.D.F: 𝑓 𝑥 = 𝜆𝑒 ;𝜆𝑥 , 𝑥 > 0
∞ 1
• Mean, 𝐸 𝑋 = 𝑅𝑋
𝑥 𝑓 𝑥 𝑑𝑥 = 0
𝑥. 𝜆𝑒 ;𝜆𝑥 𝑑𝑥 = 𝜆
∞ 2 2
• 𝐸 𝑋2 = 0
𝑥 . 𝜆𝑒 ;𝜆𝑥 𝑑𝑥 = 𝜆2
1
• Variance, 𝑣𝑎𝑟 𝑥 = 𝐸 𝑋 2 − 𝐸 𝑋 2 = 𝜆2
∞ 𝑡𝑥 𝜆
• MGF, 𝑀𝑋 𝑡 = 0
𝑒 𝜆𝑒 ;𝜆𝑥 𝑑𝑥 = 𝜆;𝑡

• Memoryless property of Exponential Distribution: 𝑃 𝑋 > 𝑠 + 𝑡 𝑋 > 𝑠 = 𝑃(𝑋 > 𝑡)

 Problem
1. The lifetime of a battery is exponential distribution with mean 4 weeks, (i) what is the probability that the
battery life exceeds 2 weeks? (ii) Given that the battery has lasted for 6 weeks, what is the probability that it
will last for at least another 5 weeks?
Solution:
Given 𝑋~exp 𝜆 & 𝑋 denotes life time of the battery
𝑓 𝑥 = 𝜆𝑒 ;𝜆𝑥 , 𝑥 > 0, 𝜆 = 0
1
𝐸 𝑥 = =4
𝜆
1
⇒𝜆=
4
1 ;𝑥
∴ 𝑓𝑋 𝑥 = 4
𝑒 4, 𝑥 ≥0
∞ 1 ;𝑥 1 1
(i) 𝑃 𝑋 > 2 = 2 4
𝑒 4 𝑑𝑥 = − 𝑒 ;∞ − 𝑒 ;2 = 𝑒 ;2 = 0.6065
𝑥 5
∞1 ;4 ;4
(ii) By memoryless property, 𝑃 𝑋 > 11 𝑋 > 6 = 𝑃 𝑋 > 5 = 5 4
𝑒 𝑑𝑥 = 𝑒 = 0.2865
2. Suppose the duration of telephone calls handled by an exchange can be reasonably assumed to have an
1
exponential distribution with parameter 𝜆 = 2 time being measured in minutes. What is the probability
that a telephone call would last at least 2 minutes, and that a call which has lasted 2 minutes lasts further 2
minutes?
Solution:
Given 𝑋~exp 𝜆 & 𝑋 denotes duration of telephone call
𝑥
−
𝑒 2
𝑓 𝑥 = 2
,𝑥 >0
∞ 1 ;𝑥
(i) 𝑃 𝑋 > 2 = 2 2
𝑒 2 𝑑𝑥 = 𝑒 ;1
𝑥
∞1 ;2
(ii) By memoryless property, 𝑃 𝑋 > 4 𝑋 > 2 = 𝑃 𝑋 > 2 = 2 2
𝑒 𝑑𝑥 = 𝑒 ;1
3. The milage which car owners get with a certain kind of radial tire is a RV having an
exponential distribution with mean 40,000 𝑘𝑚. Find the probabilities that one of these tires
will last (i) at least 20,000 km and (ii) at most 30,000 km.
Solution:
𝑥
1 ;
𝑓 𝑥 = . 𝑒 40000 , 𝑥 >0
40000
𝑥
∞ 1 ;40000
(i) 𝑃 𝑋 ≥ 20000 = 20000 40000
𝑒 𝑑𝑥 = 𝑒 ;0.5 = 0.6065
𝑥
30000 1 ;40000
(ii) 𝑃 𝑋 ≤ 30000 = 0 40000
𝑒 𝑑𝑥 = 1 − 𝑒 ;0.75 = 0.5270
 Normal Distribution
𝑥−𝜇 2
1 ;
• 𝑋~𝑁(𝜇, 𝜎 2 ) , 𝑓 𝑥 = 𝜎 2𝜋
𝑒 2𝜎2 , −∞ < 𝑥 < ∞

Remark:
• If 𝑋~𝑁(𝜇, 𝜎 2 ), then curve is a bell shaped curve & symmetric along 𝑥 = 𝜇
• Any random variable shifted with 𝜇 & scale it by 𝜎 then it is standard normal distribution.
𝑧 2
𝑋;𝜇 1 ;2
• 𝑍= 𝜎
~𝑁 0, 1 , 𝑓 𝑧 = 2𝜋
𝑒 , −∞ < 𝑧 < ∞
𝜎2𝑡2
• 𝑀𝑋 𝑡 = 𝑒 𝜇𝑡: 2 , 𝐸 𝑋 = 𝜇, 𝑉𝑎𝑟 𝑋 = 𝜎 2
 Problem
1. The mean weight of 500 students is 151 pounds and S.D is 15 pounds. Assuming that the weights are
normally distributed. Find how many students weight (i) between 120lb &155lb (ii) more than 185lb
Solution:
𝑋-Weights of students~𝑁(151, 152 )
Given 𝜇 = 151, 𝜎 = 15
120;𝜇 𝑋;𝜇 155;𝜇
(i) 𝑃 120 < 𝑋 < 155 = 𝑃 𝜎
<
𝜎
<
𝜎
= 𝑃(−2.067 < 𝑧 < 0.267)
= 𝐹 0.267 − 𝐹 −2.067
= 𝐹 0.27 − 𝐹(−2.07)
= 0.6064 − 0.0192
= 0.5872
No. of students between 120lb & 155lb weight is 0.5872× 500 = 294
185;𝜇
(ii) 𝑃 𝑋 > 185 = 𝑃 𝑍 > = 𝑃 𝑍 > 2.27 = 1 − 𝐹 2.27 = 0.0116
𝜎
No. of students whose weight is more than 185lb is 0.0116 × 500 = 6
• In a class of 50, the average mark of students in a subject is 48 & standard
deviation is 24. Find the no. of students who get (i) above 50 (ii) between 35 & 50?
Solution
𝑋~𝑁(48, 576)
𝐺𝑖𝑣𝑒𝑛 𝜇 = 48, 𝜎 = 24, 𝜎 2 = 576
𝑋;𝜇 50;𝜇
(i) 𝑃 𝑋 > 50 = 𝑃 𝜎
> = 𝑃 𝑍 > 0.083 = 1 − 𝐹 0.083
𝜎
= 1 − 0.5319 = 0.4681
No. of students got above 50 marks = 0.4681 × 50 = 23
35;𝜇 50;𝜇
(ii) 𝑃 35 < 𝑋 < 50 = 𝑃 <𝑧< = 𝑃 −0.54 < 𝑋 < 0.08
𝜎 𝜎
= 𝐹 0.08 − 𝐹 −0.54 = 0.5319 − 0.2946 = 0.2373
No. of students who got marks between 35 to 50 = 12
• The heights of a student in the class is normally distributed, 6% have height below
60 inches & 39% are between 60 & 70 inches. Find the mean & SD of the height?
Solution:
𝑋 denotes height of the students, 𝑋~𝑁(𝜇, 𝜎)
Given 𝑃 𝑋 < 60 = 0.06, 𝑃 60 < 𝑋 < 70 = 0.39
60;𝜇
𝑃 𝑍< = 0.06 ------(1)
𝜎
60;𝜇 70;𝜇
𝑃 <𝑍< = 0.39 -----(2)
𝜎 𝜎
60;𝜇
Consider 𝑃 𝑍 < = 0.06
𝜎
60;𝜇
⇒ 𝐹 = 0.06
𝜎
60;𝜇
⇒ = −1.55 -----(3)
𝜎
 60;𝜇 70;𝜇
Similarly, 𝑃 <𝑍< = 0.39
𝜎 𝜎
70;𝜇 60;𝜇
⇒𝐹 −𝐹 = 0.39
𝜎 𝜎
70;𝜇
⇒𝐹 − 0.06 = 0.39
𝜎
70;𝜇
⇒𝐹 = 0.45
𝜎
70;𝜇
⇒ = −0.13 ----(4)
𝜎
Solving (3) and (4) we get
𝜇 = 70.91 ≈ 71
𝜎 = 7.04 ≈ 7
 Gamma or Erlang Distribution
• 𝑋~𝐺𝑎𝑚𝑚𝑎(𝑘, 𝜆) where k ≥ 0, 𝜆 ≥ 0
𝜆𝑒 −𝜆𝑥 𝜆𝑥 𝑘−1
• 𝑓𝑥 = , 𝑥 ≥ 0 , Where Γ 𝑘 = ∞ ;𝑥 𝑘;1
𝑒 𝑥 𝑑𝑥 or Γ 𝑘 = 𝑘 − 1 !
Γ𝑘
0
0, 𝑜𝑡𝑕𝑒𝑟𝑤𝑖𝑠𝑒
• In particular, if k = 1, 𝑓 𝑥 = 𝜆𝑒 ;𝜆𝑥 , 𝜆 ≥ 0, 𝑥 ≥ 0 which is an exponential distribution.
 𝛼
Mean, 𝐸 𝑋 =
𝜆
𝛼 𝛼:1
𝐸 𝑋2 = 𝜆2
𝛼
Variance, 𝑣𝑎𝑟 𝑋 = 𝐸 𝑋 2 − 𝐸 𝑋 2 = 𝜆2

𝑡 ;𝛼
MGF, 𝑀𝑋 𝑡 = 1 − 𝜆
 Problem
1. In a certain city, the daily consumption of electric power in millions of Kilowatt-hours can be treated as a
1
RV having an Erlang distribution with parameters 𝜆 = 2 and k = 3. If the power plant of this city has a
daily capacity of 12 millions kilowatt-hours. What is the probability that this power supply will be
inadequate on any given day.
Solution:
Let 𝑋 represent the daily consumption of electric power.
1
𝑋~𝐸𝑟𝑙𝑎𝑛𝑔(𝜆 = ,k = 3)
2
𝑥
𝜆𝑘 𝑒 −𝜆𝑥 𝑥 𝑘−1 1 3 − 2
,𝑥 ≥ 0 = 2
𝑒 2𝑥
𝑓 𝑥 = Γ𝑘
Γ(3)
,𝑥 ≥ 0
0, 𝑜𝑡𝑕𝑒𝑟𝑤𝑖𝑠𝑒
∞
𝑃 𝑇𝑕𝑒 𝑝𝑜𝑤𝑒𝑟 𝑠𝑢𝑝𝑝𝑙𝑦 𝑖𝑠 𝑖𝑛𝑎𝑑𝑒𝑞𝑢𝑎𝑡𝑒 = 𝑃 𝑋 > 12 = 12
𝑓(𝑥) 𝑑𝑥
∞ ;𝑥 2 ∞
Γ 3 = 0
𝑒 𝑥 𝑑𝑥 = −𝑥 2 𝑒 ;𝑥 + 2𝑥𝑒 ;𝑥 − 2𝑒 ;𝑥 0 = 0 − −2 =2
1 ∞ ;𝑥 2
𝑃 𝑋 > 12 = 16 12
𝑒 2 𝑥 𝑑𝑥 = 0.0625
2. If a R.V 𝑋 has a Erlang Distribution with 𝑘 = 2, 𝜆 = 1. Find 𝑃 1.8 < 𝑋 < 2.4 .
Solution:
𝑋~𝐸𝑟𝑙𝑎𝑛𝑔(2, 1)
𝑒 −𝑥 𝑥
,𝑥 ≥0
𝑓 𝑥 = Γ(2)
0, 𝑜𝑡𝑕𝑒𝑟𝑤𝑖𝑠𝑒
2.4 ;𝑥
𝑃 1.8 < 𝑋 < 2.4 = 1.8
𝑥𝑒 𝑑𝑥 = 0.1545
3. Let 𝑋 be a R.V having Erlang distribution with mean 6 and variance 12. find the probability
that 𝑃 𝑋 ≥ 12 ?
Solution:
𝑘
Mean = =6
𝜆
𝑘
Variance = = 12
𝜆2
1
⇒ 𝜆 = ,𝑘 = 3
2

1 ;𝑥 𝑥 2
𝑓 𝑥 = 𝑒 2. ,𝑥 ≥0
2 2

∞ 1 ;𝑥 𝑥 2
𝑃 𝑋 ≥ 12 = 12 2
𝑒 2. 𝑑𝑥 = 0.06196
2
4. If a company employees 𝑛 sales person, its gross sales in thousands of rupees may be regarded as a RV
1
having an Erlang distribution with 𝜆 = 2 and k = 80,000 𝑛. If the sales cost is Rs. 8000 per salesperson, how
many salespersons should the company employ to maximise the expected profit?
Solution:
Let 𝑋 represent the gross sales (in thousand of rupees) by 𝑛 sales person
1
𝑋~𝐸𝑟𝑙𝑎𝑛𝑔 with 𝜆 = and k = 80000 𝑛.
2
𝑘
𝐸 𝑋 = 𝜆 = 1,60,000 𝑛
If y denote the total expected profit of the company, then, 𝑦 = 𝑡𝑜𝑡𝑎𝑙 𝑒𝑥𝑝𝑒𝑐𝑡𝑒𝑑 𝑠𝑎𝑙𝑒𝑠 − 𝑡𝑜𝑡𝑎𝑙 𝑠𝑎𝑙𝑒𝑠 𝑐𝑜𝑠𝑡
𝑦 = 1,60,000 𝑛 − 8000 𝑛
𝑑𝑦 80,000
𝑑𝑛
= 𝑛
− 8000
𝑑𝑦
= 0 => 𝑛 = 10 or 𝑛 = 100
𝑑𝑛
𝑑2 𝑦
When 𝑛 = 100, 𝑑𝑛2
<0
Therefore y is maximum, when 𝑛 = 100
That is company should employ 100 salespersons in order to maximise the total expected profit.
 Weibull Distribution
• A continuous RV X is said to follow a Weibull Distribution with parameters
𝛼, 𝛽 > 0, if the RV 𝑌 = 𝛼𝑋𝛽 follows the exponential distribution with density
function 𝑓𝑌 𝑦 = 𝑒 ;𝑦 , 𝑦 > 0
• Density function of the Weibull Distribution
𝛽
• 𝑓𝑋 𝑥 = 𝛼𝛽𝑥 𝛽;1 . 𝑒 ;𝛼𝑥 ; 𝑥 > 0

• Note: When 𝛽 = 1, Weibull distribution reduces to the exponential distribution

with parameter 𝛼
 Weibull Distribution
1
;𝛽 1
• Mean, 𝐸 𝑋 = 𝛼 . Γ 𝛽
+1

2
2
;𝛽 2 1
• Variance = 𝛼 . Γ 𝛽
+1 − Γ
𝛽
+1
 Problem
1. Each of the 6 tubes of a radio set has a life length (in years) which many be considered as a RV that follows a Weibull
distribution with parameters 𝛼 = 25 and 𝛽 = 2. If these tubes function independently of one another, what is the
probability that no tube will have to be replaced during the first 2 months of service?
Solution:
If 𝑋 represents the life length of each tube, then its density function
𝛽
𝑓 𝑥 = 𝛼𝛽𝑥 𝛽;1 . 𝑒 ;𝛼𝑥 ; 𝑥 > 0
2
𝑓 𝑥 = 50𝑥. 𝑒 ;25𝑥 ; 𝑥 > 0
1
𝑃(a tube is not to be replaced during the first 2 months of service)= 𝑃(𝑋 > )
6
1 ∞ 2
𝑃 𝑋> = 1 50𝑥. 𝑒 ;25𝑥 𝑑𝑥
6 6
1 1
Put 𝑥 2 = 𝑡 ⇒ 2𝑥𝑑𝑥 = 𝑑𝑡 when 𝑥 = then 𝑡 = , when 𝑥 = ∞, 𝑡𝑕𝑒𝑛 𝑡 = ∞
6 36
1 ∞
𝑃 𝑋> = 1 25. 𝑒 ;25𝑡 𝑑t
6 36
∞ 25
1 𝑒 −25𝑡
𝑃 𝑋> = 25. = 𝑒 ;36
6 ;25 1/36
25 25
; 6 ;
P(All the 6 tubes are not to be replaced during the first 2 months)= (𝑒 36 ) =𝑒 6 (By independence)
2. If the life 𝑋 (in years) of a certain type of car has a Weibull distribution with the parameter
𝛽 = 2, find the value of the parameter 𝛼, given the probability that the life of a car exceeds 5
years is 𝑒 ;0.25 . For these values of 𝛼 and 𝛽, find the mean and variance of 𝑋
Solution:
𝛽
The density function of 𝑋 is given by 𝑓 𝑥 = 𝛼𝛽𝑥 𝛽;1 . 𝑒 ;𝛼𝑥 ; 𝑥 > 0
2
𝑓 𝑥 = 2𝛼𝑥𝑒 ;𝛼𝑥 , 𝑥 > 0
𝑃 𝑋 > 5 = 𝑒 ;0.25
∞ ;𝛼𝑥 2
⇒ 5
2𝛼𝑥𝑒 𝑑𝑥 = 𝑒 ;0.25
;25𝛼 ;0.25
⇒𝑒 =𝑒
1
⇒ 𝛼=
100
𝑥2
2 ;100
Therefore 𝑓 𝑥 = 𝑥𝑒 ,𝑥 >0
100
 1
; 1
Mean = 𝛼 𝛽 .Γ +1
𝛽
1
1 ;2 3 1 1
𝑀𝑒𝑎𝑛 = 100
. Γ 2
= 10 × 2 Γ 2
=5 𝜋

2 2
;𝛽 2 1
Variance = 𝛼 . Γ +1 − Γ +1
𝛽 𝛽

𝜋
Variance =100(1 − 4 )
3. If the time 𝑇 to failure of a component follows a Weibull distribution with
parameter 𝛼 and 𝛽, find the hazard rate or conditional failure rate at time 𝑡 of the
component.
Solution:
If 𝑓 𝑡 is density function of 𝑇 and 𝑕(𝑡) is the hazard rate at time t, then
𝑓 𝑡
𝑕 𝑡 =
1;𝐹(𝑡)
𝛽;1 ;𝛼𝑡 𝛽
Here 𝑓 𝑡 = 𝛼𝛽 𝑡 𝑒 ,𝑡 >0
𝑡 𝑡
𝐹 𝑡 =𝑃 𝑇≤𝑡 = 𝛼𝛽 𝑡 𝛽;1 𝑒 ;𝛼𝑡 𝛽 𝑑𝑡 = −𝑒 ;𝛼𝑡
𝛽
= 1 − 𝑒 ;𝛼𝑡
𝛽
0 0
𝛽
𝛼𝛽 𝑡 𝛽−1 𝑒 −𝛼𝑡
𝑕 𝑡 = 𝛽 = 𝛼𝛽 𝑡𝛽;1
𝑒 −𝛼𝑡