PROBABILITY AND STATISTICS

Chapter 3 (3.4-3.6): Binomial-Hypergeomtric

Negative Binomial and Poisson PD

Examples

Exam0.1. What score a monkey get by taking the test.

Problem:
A test consists of 50 multiple choices questions. Each question has 4 choices and only one
choice is correct. Suppose one just take the test by randomly anwser without even reading.
1) What is the probability of having at least 20 correct answers on this test (score at least
4/10)?
The same for 15 or 10 correct anwsers (score at least 3/10 or 2/10).
2) Let's consider the same problem if each question has 5 choices.
3) Find out the formula to compute the probability P X = x .
Sol: Compute directly
This is a binomial distribution with size 50 and p = 0.25. Then we can sea
P X R 20 = 1 KP X % 19 .
O p d 0.20; x d 20; n d 50;
p := 0.20
x := 20
n := 50 (1.1.1.1.1)
i n Ki
O 1 K sum binomial n, i $p $ 1 Kp , i = 0 ..x K1
0.0009324365 (1.1.1.1.2)
O
Note that it is hard to compute without software. The same computation with R.

or
O
It is more than 25 %.
If each question has 5 choices
 O
O

Exam0.2 (Earthquakes)
Problem:
For a recent period of 100 years (from 1900 to 2000), there were 93 major earthquakes
(measuring at least 6.0 on the Richter scale) in the World (based on data from the World
Almanac and Book of Facts).
How can we compute:
a. The average number of major earthquakes per year ?
b. The probability P x of having x earthquakes in a randomly selected year, find
P x for x = 0 to 10.
Sol:
The problem can be considered as Poisson distribution.
O restart :
Ka x
exp 1 $a
O Ps d ;
x!
Ka x
e a
Ps := (1.2.1.1.1)
x!
O
O subs x = 0, a = 0.93, Ps : evalf %
0.3945537104 (1.2.1.1.2)
O subs x = 1, a = 0.93, Ps : evalf %
0.3669349507 (1.2.1.1.3)
O subs x = 2, a = 0.93, Ps : evalf %
0.1706247521 (1.2.1.1.4)
O subs x = 4, a = 0.93, Ps : evalf %
0.01229777900 (1.2.1.1.5)

Binomial Probability Distribution

Definition
O

Examples
Exam 3.27

O
Exam 3.29

Q: The experiment is a binomial?

sol
 O
O
Exam 3.30 (approx. Binomial)

O
O

Rule

The Binomial Random Variable and Distribution

 Definition

O
Notation

O
Works: Find down pmf of b x; n, p
n=2,3
n=4
O
Formula-Theorem

CDF for Binomial RV-Table of CDF (Apendix A.1)

O

Exam 3.32

Apendix Table A.1

O
O

The Mean and Variance of X

Proposition

O
Exam3.34

Maple
O b d binomial n, x $px$ 1 Kp nKx;
x n Kx
b := binomial n, x p 1 Kp (2.8.1)
O subs x = 7, n = 10, p = 0.75, b : evalf %
0.2502822876 (2.8.2)
O subs x = 7, n = 10, p = 0.75, b Csubs x = 8, n = 10, p = 0.75, b : evalf %
0.5318498611 (2.8.3)
O binomial 10, 7 #to compute combinations in Maple
120 (2.8.4)
O with combinat, numbperm :#You can compute permuatations now
O numbperm 10, 7
604800 (2.8.5)

R Script for binomial PD

The Hypergeometric Distribution
The hypergeometric and negative binomial distributions are both related to the binomial distribution.

Characteristic
O

Exams
Exam 3.35

Formula
Proposition

O
Exam 3.36
 O

Expected value and variance of X

Proposition

O
Note

O
Exam 3.37
 O

Maple
binomial M, x $binomial N KM, n Kx
O hd ;
binomial N, n
binomial M, x binomial N KM, n Kx
h := (3.5.1)
binomial N, n
O subs x = 2, n = 3, N = 7, M = 4, h ; evalf %
binomial 4, 2 binomial 3, 1
binomial 7, 3
0.5142857143 (3.5.2)
5
O >subs
x= 3
n = 7, N = 15, M = 5., h ;

0.4265734266 (3.5.3)

The Negative Binomial Distribution

Characteristic
The negative binomial rv and distribution are based on an experiment satisfying the
following conditions:
Counting-Formula

Proposition

(4.2.1.1)

Exam 3.38

Note (Geometric distribution again)

 Expected value and Variance
Proposition

Maple
O nb d binomial x Cr K1, r K1 $pr$ 1 Kp x;
r x
nb := binomial x Cr K1, r K1 p 1 Kp (4.6.1)
O subs x = 2, r = 3, p = 0.7, nb ; evalf %
0.03087 binomial 4, 2
0.18522 (4.6.2)
10
O >binomial x C5 K1, 5 K1 $0.2 0.8 ;
x= 0
5 x

Poisson Probability Distribution

Definition
• The Poisson distribution is a discrete probability distribution that applies to occurrences of
some event over a specified interval.
• The random variable x is the number of occurrences of the event in an interval.
• The interval can be time, distance, area, volume, or some similar unit.

where l is the average of occurences in the specified interval.

P 0 = 0.394
 Remark: Requirements for the Poisson Distribution

Mean and Variance

Proposition

Proof: Compute directly by using the following formula.

Examples
Exam 3.39

Exam (Earthquakes)
 For a recent period of 100 years (from 1900 to 2000), there were 93 major earthquakes
(measuring at least 6.0 on the Richter scale) in the World (based on data from the World
Almanac and Book of Facts).
How can we compute:
a. The average number of major earthquakes per year ?
b. The probability P x of having x earthquakes in a randomly selected year, find
P x for x = 0 to 10.
c. The actual results are as follows: 47 years (0 major earthquakes); 31 years (1 major
earthquake); 13 years (2 major earthquakes); 5 years (3 major earthquakes); 2 years (4 major
earthquakes); 0 years (5 major earthquakes); 1 year (6 major earthquakes); 1 year (7 major
earthquakes). How do these actual results compare to the probabilities found in part b.? Does
the Poisson distribution appear to be a good model in this case?

Sol:

The Poisson Distribution as a Limit

Proposition

O
Exam 3.40
 Comparison

O
Exam 3.42

Maple
exp 1 Ka$ax
O Ps d :
x!
O subs x = 0, a = 0.93, Ps : evalf %
(5.5.1)
 0.3945537104 (5.5.1)
O 1Ksubs x = 0, a = 4.6, Ps : evalf %
0.9899481642 (5.5.2)
96 C480 656.
O C
96 C480 C656 C2330 96 C480 C656 C2330
0.3458731050 (5.5.3)

4 4
O $
52 51.
0.006033182504 (5.5.4)
O

Homework:
You can use R for your checking of solutions

3.4
I. 46,47,48
II. 49-55 al. 2; 56-60 al. 2
III. 61-67 al. 2
O

3.5
I. 68, 69, 71
III. 70,72-78 al 2
O

3.6
I. 79-81
II. 82-85 al. 2
III. 86-93 al. 4
O