16/03/2023

SOME COMMON
CHAPTER 3 PROBABILITY DISTRIBUTIONS
Types of Distributions
SOME COMMON  Bernoulli Distribution
PROBABILITY DISTRIBUTIONS  Binomial Distribution

 Poisson Distribution

 Normal Distribution
Dr. Lam Son
[email protected]

 Uniform Distribution
033.6969.909

 Exponential Distribution

Bernoulli Distribution Examples of Bernoulli distribution

 The experiment consists of a single trial with two possible outcomes
 it’s going to rain tomorrow or not?
 only two possible outcomes: A rain denotes success and no rain denotes failure.
1 (success) 0 (failure)  How many boys are born each day?
p(success)=p p(failure)=1-p - You want to find out how many boys are born each
X = 1 if the outcome is success day, so you call a boy birth a “success” and a girl
= 0 if the outcome is failure birth a “failure.”
 X has a Bernoulli distribution.  The probability of a roll of two die resulting in a
The probability mass function is given by: double six. A double six die roll would be your
“success” and everything else rolled would be
P(x) = px (1-p)1-x where x = 0; 1.
considered a “failure.”
E(X) = p; V(X) = p(1-p)

Bernoulli Trial Independence

 A Bernoulli trial is one of the simplest experiments  Each trial must be independent: the outcome of
you can conduct in probability and statistics. It’s an one trial does not affect the outcomes of any
experiment where you can have one of two possible other trials.
outcomes. For example, “Yes” and “No” or “Heads”
and “Tails.”
 A series of Bernoulli trials:
 Bernoulli trials are usually phrased in terms - Each trial is independent.
of success and failure. Success doesn’t mean
success in the usual way — it just refers to an - Each trial has only two possible outcomes:
outcome you want to keep track of. 1 (success) and 0 (failure)
p(success)=p p(failure)=1-p

1
 16/03/2023

Binomial Distribution Binomial Random Variable

 Binomial Experiment  The binomial random variable is the number of

- Each trial is independent. successes in the Binominal experiment’s n
- There are only two possible outcomes in a trial- trials.
either a success or a failure. It can take on values 0, 1, 2, . . . , n.
- Total number of trials: n
- The probability of success and failure is same for all Thus, the random variable is discrete.
trials. (Trials are identical).

Binomial Probability Distribution Example

Pat Statsdud is a student taking a statistics course.
 The probability of x successes in a binomial Unfortunately, Pat is not a good student. Pat does not read
experiment with n trials and probability of the textbook before class, does not do homework, and
success = p is regularly misses class. Pat intends to rely on luck to pass
the next quiz. The quiz consists of 10 multiple-choice
questions. Each question has five possible answers, only
p (k ) = P( X = k ) = Cnk ⋅ p k (1 − p) n − k one of which is correct. Pat plans to guess the answer to
for k=0,1,2,…,n. each question.
a. What is the probability that Pat gets no answers correct?
E(X) = np, V(X) = np(1-p)
b. What is the probability that Pat gets two answers correct?
np+p-1 ≤ mod(X) ≤ np+p
c.

Example Excel

 The experiment is binomial with n =10 and p = 0.2  Type the following into any empty cell:
Let X be the number of correct answers in n trials.
X~B(10; 0.2) =BINOMDIST([k], [n], [p], [True] or [False])
a) P(X=0) =?
b) P(X=2) = ? Typing “True” calculates a cumulative probability and
c) Find the probability that Pat fails the quiz. A typing “False” computes the probability of an
mark is considered a failure if it is less than 50%. individual value of X.
P(X ≤ 4) = ?

2
 16/03/2023

Poisson distribution Poisson Experiment

 The number of cars arriving at a service station in 1
1. The number of successes that occur in any interval is
hour. (The interval of time is 1 hour.)
independent of the number of successes that occur in
any other interval.
 The number of flaws in a bolt of cloth. 2. The probability of a success in an interval is the same
(The specific region is a bolt of cloth.) for all equal-size intervals.
3. The probability of a success in an interval is
 The number of accidents in 1 day on a particular proportional to the size of the interval.
stretch of highway. (The interval is defined by both 4. The probability of more than one success in an
time, 1 day, and space, the particular stretch of interval approaches 0 as the interval becomes smaller.
highway.)

Poisson Random Variable Example

 The Poisson random variable is the number of  A statistics instructor has observed that the number of
successes that occur in a period of time or an interval typographical errors in new editions of textbooks varies
considerably from book to book. After some analysis, he
of space in a Poisson experiment.
concludes that the number of errors is Poisson distributed
µx with a mean of 1.5 per 100 pages. The instructor randomly
P( X = x) = p( x) = e−µ . selects 100 pages of a new book. What is the probability
x! for x = 0; 1; 2; …
that there are no typographical errors?
X 0 1 2 .. x
 Suppose that the instructor has just received a copy of a
new statistics book. He notices that there are 400 pages.
2
µx
e −µ
P
e −µ .µ e−µ . µ e −µ . a. What is the probability that there are no typos?
2 x!
b. What is the probability that there are five or fewer typos?

Using Excel µ,σ

Normal Distribution N(µ σ2)
 The probability density function of a normal random
variable is: 2
1  x −µ 
1 −  
f ( x) = ⋅e 2 σ 
−∞ < x < ∞
σ 2π

E(X)=µ
V(X)=σ2

3
 16/03/2023

Normal Distributions with the Same Variance Normal Distributions with the Same Means
but Different Means but Different Variances

Standard Normal Random Variable P (-0.71<Z<0.92)

X −µ
X ∼ N (µ; σ 2 )  Z = ∼ N ( 0;1)

σ
z2
1 −
f ( z) = ⋅e 2

2π

Finding Values of Z Z0.025=1.96 ⇔ p(Z>1.96)=0.025

 P(Z > ZA) = A

4
 16/03/2023

Because the standard normal curve is EXCEL: To compute

symmetric about 0  -ZA? cumulative normal probabilities

 X∼N(µ,σ2): p(X≤x)= NORMDIST([X], [µ], [σ],True)

x −µ
 P(X<x) = p(Z< = zA) =
σ
NORMSDIST(zA)=F(zA)

 NORMSINV([1 – A]) = zA

z1-A = - zA

X −µ X −µ x−µ x−µ
Practice Z= P( X < x) = P( < ) = P(Z < )
σ σ σ σ

 8.15. p(Z<1.50) = 0.9332

 8.18. p(Z<-1.59) = P(Z>1.59)= 1-P(Z<1.59)=1-0.9441 = 8.34. X is normally distributed with mean 100 and
0.0559 standard deviation 20. What is the probability that
 8.21. p(-1.4 < Z < 0.60) = P(Z<0.6)-P(Z<-1.4)= 0.6449 X is greater than 145?
 8.29. p(Z>0) = 0.5=P(Z<0)=1-0.5
X~N(100;202)
 8.31. z0.02 =2.05 z0.01 = 2.33; ;z0.025 =1.96 ;z0.05 =1.645
P(X>145) = P(Z>2.25) (= (145-100)/20)
 8.32. z0.1 = 1.28
=1-P(Z<2.25) = 1-0.9878 = 0.0122
 8.34. X is normally distributed with mean 100 and standard
P(X<145) = P(Z<2.25)
deviation 20. What is the probability that X is greater than
145? =0.9878
P(90<X<125) = P(-0.5<Z<1.25) =8944 – 0.3085
= P(Z<1.25) – P(Z<-0.5)

Uniform Distribution Exponential Distribution

 A variable X is said to be uniformly distributed if the  A random variable X is exponentially distributed if its
density function is probability density function is given by
 0 if x ∉ [a, b]
 λe −λx if x ≥ 0
f ( x) =  1
 b − a if x ∈ [a, b] f ( x) = 
0 if x < 0

 Mean: E(X) = 1/λ

 Variance: Var(X) = (1/λ)²

5
 16/03/2023

Probability Associated with an

Exponential Random Variable Example

 The production of a complex chemical needed for

anticancer drugs is exponentially distributed with λ
= 6 kilograms per hour. What is the probability that
the production process requires more than 15
minutes to produce the next kilogram of drugs?

Excel: P(X<x) = EXPONDIST([x],[λ], True)

3.4. - χ2(n) 3.5. Student Distribution- T(n)

n
χ 2 =  X i2
•ĐN:
U
 T= ∼ T ( n)
i =1 V
∞
n
Gamma Function: Γ ( x ) =  t x −1e − t dt E(T)=0
0

E(χ2)=n V(T)=n/(n-2)

V(χ2)=2n P(T> Tα(n))=α

f(t)
f(χ2)
P(χ2> χα2(n))=α t 1( −n α) = − t α( n )
α
α α
The picture can't be display ed.

χ 2 t 1(−nα ) 0 t α( n ) u

3.2.6. Fisher-Snedecor Distr– F(n1,n2) 3.2.6. Fisher-Snedecor Distr– F(n1,n2)

•ĐN: U n2
E(F ) =
n2 − 2
n
 F = 1 ∼ F (n1 , n2 ) 2.n22 ( n1 + n22 − 2)
V f(x) V (F ) =
n1 ( n2 − 2) 2 ( n2 − 4)
n2
U~χ2(n1), V~χ2(n2) α

fα( n1 , n2 ) x
p ( F > fα ( n1 , n2 )
) =α
1
fα( n1 , n2 ) = ( n2 , n1 )
f 1−α