Vous êtes sur la page 1sur 28

SIT1001 PROBABILITY & STATISTICS

NEGATIVE BINOMIAL
DISTRIBUTION
2

HELLO!
We are …
1.Anatasha binti Ahmad Nor (SIT180003)
2.Chor Suet Yeen (SIQ180009)
3.Haikal Irfan bin Azuan Aziz Kasim (SIQ180010)
4.Mathila Lizanne James (SIT180008)
5.Ng Yi Quan (SIQ180025)
6.Nur Siti Zulaikha bt Jailani (SIM160031)
3

1.
INTRODUCTION
- Standard denotation of distribution
4

INTRODUCTION
» The negative binomial distribution is the
distribution of the number of trials needed to
GET a fixed number of successes.
» How to determine an event is a negative binomial
distribution or not?
CHARACTERISTICS of
5

Negative Binomial Distribution


» The experiment consists of x repeated trials.
» Each trial can result in just TWO possible
outcomes, a success and a failure.
» The probability of success, denoted by p, is the
same on every trial.
» The trials are INDEPENDENT; that is, the outcome
on one trial does not affect the outcome on other
trials.
» The experiment continues until r successes are
observed, where r is specified in advance.
6

Notation and Parameter


» X ~ NB(r,p)
PARAMETER DESCRIPTION

r Number of successes in the negative binomial


binomial experiment
p Probability of success in each experiment
7

2.
PROPERTIES of distribution
- pmf - Mean
- cdf - Variance
- mgf
8

Probability mass function (pmf)


» Negative binomial distribution is DISCRETE
random variable
» The probability mass function is :
𝑥−1 𝑟
» 𝑃 𝑋=𝑥 = 𝑝 1 − 𝑝 𝑥−𝑟
𝑟−1
• P(success)=p, this stay constant from trial to
trial
• P(failure)=1-p
• X represents the trial number of the rth success
9

Derivation of pmf
For the rth success to occur on the xth trial:
» The first x-1 trials must result in r-1 successes
» Thus by using binomial formula:
𝑥 − 1 𝑟−1
𝑝 1 − 𝑝 𝑥−1 − 𝑟−1
𝑟−1
» AND The xth trial must be a success, which has a
probability of p
» Thus, by multiply BOTH together we will get the
pmf
10

Cont…
» The probability the rth success occurs on the xth
trial is:

𝑥 − 1 𝑟−1 𝑥−𝑟
𝑃 𝑋 =𝑥 =𝑝× 𝑝 1−𝑝
𝑟−1
𝑥−1 𝑟
𝑃 𝑋=𝑥 = 𝑝 1 − 𝑝 𝑥−𝑟
𝑟−1
11

Cumulative distribution function (CDF)



𝑋−1
෍) (𝑝𝑟 ) 1 − 𝑝 𝑋−𝑟 =1
𝑟−1
𝑋=𝑟
12

Moment generating function (mgf)


13

Mean, µ / Expected value


From the mgf,

𝑀′ 𝑡 = 𝑝et 𝑟
−𝑟 1 − 1 − 𝑝 𝑒 𝑡 −𝑟−1
− 1 − 𝑝 𝑒𝑡

+ 𝑟 𝑝et 𝑟−1 (𝑝𝑒 𝑡 )[1 − 1 − 𝑝 𝑒 𝑡 ](−𝑟)

= 𝑟 𝑝et 𝑟 [1 − 1 − 𝑝 𝑒 𝑡 ](−𝑟−1)
14

Mean, µ / Expected value


Therefore, the expected value for the negative binomial distribution will be;
𝐸 𝑋 = 𝑀′ 0
= 𝑟𝑝𝑟 𝑝(−𝑟−1)
= 𝑟𝑝−1
𝑟
=
𝑝
15

Variance, Var(X)
From the mgf,

𝑀′′ 𝑡 = 𝑟 𝑝et 𝑟
−𝑟 − 1 1 − 1 − 𝑝 𝑒 𝑡 −𝑟−2
− 1 − 𝑝 𝑒𝑡

+ 𝑟 2 𝑝et 𝑟−1 (𝑝𝑒 𝑡 )[1 − 1 − 𝑝 𝑒 𝑡 ](−𝑟−1)


16

Variance, Var(X)
» Therefore, the variance for the negative binomial distribution will be:

𝑉𝑎𝑟 𝑋 = 𝑀′′ 0 − [𝑀′ 0 ]2


𝑟 2
= 𝑟 𝑟 + 1 𝑝𝑟 𝑝 −𝑟−2
1 − 𝑝 + 𝑟 2 𝑝𝑟 𝑝 −𝑟−1

𝑝

𝑟
= 𝑟𝑝−2 1 − 𝑝 𝑟 + 1 + 𝑟𝑝 − 2
𝑝
17

Variance, Var(X)
𝑟(𝑟 + 1 − 𝑝) 𝑟 2
𝑉𝑎𝑟 𝑋 = −
𝑝2 𝑝
𝑟(𝑟+1−𝑝) 𝑟 2
= −
𝑝2 𝑝
𝑟(1−𝑝)
=
𝑝2
𝑟𝑞
=
𝑝2
18

3.
SHAPEs of distribution
19

r=3, p=0.7

V1 V2 V3 V4 V5 V6
(set seed 10) (set seed 20) (set seed 50) (set seed 100) (set seed 999) (Set Seed 1000)
20

Set seed=1000

V1 V2 V3 V4 V5 V6
(r=3,P=0.7) (r=4,P=0.7) (r=10,P=0.7) (r=15,P=0.7) (r=50,P=0.7) (r=100,P=0.7)
21

Set seed=1000

V1 V2 V3 V4 V5 V6
(r=3,P=0.1) (r=3,V=0.2) (r=3,P=0.5) (r=3,P=0.75) (r=3,P=0.9) (r=3,p=0.99)
22

4.
APPLICATIONS
23

QUESTION 1
» Recently it is found that the bottom of the Baltic
r=5
Sea has potential of oil discovery. Suppose that a
n=9-5=4
well oil drilling has 18% chance of striking oil. Find p=0.18
the probability that the fifth oil strike comes on the q=0.82
9 well drilled.
» SOLUTION:
24

QUESTION 2
» Supposed that the probability of a child exposed to
r=4
the flu will catch the flu is 0.4. What is the
n=12-4=8
probability that the twelfth child exposed to the flu p=0.4
will be the fourth to catch it? q=0.6
» SOLUTION:
25

QUESTION 3a
» An oil company conducts a geological study that
r=1
indicates that an exploratory oil well should have a
n=3-1=2
20% chance of striking oil. What is the probability p=0.2
that the first strike comes on the third well drilled? q=0.8
» SOLUTION:
3−1
P(X=3) = (1−p)3−1 p1 = (1−p)2p = 0.802×0.20
1−1
= 0.128
26

QUESTION 3b
» What is the probability that the third strike comes
r=3
on the seventh well drilled?
n=7-3=4
» SOLUTION: p=0.2
7−1 6 q=0.8
P(X=7) = (1−p)7−3 p3 = 0.804×0.203
3−1 2
= 0.049
27

QUESTION 3c
» What is the mean and variance of the number of
wells that must be drilled if the oil company wants
to set up three producing wells?
» SOLUTION:
μ = E(X) = r/p = 3/0.20 = 15
σ2 = Var(x) = r(1−p)/p2 = 3(0.80)/0.202 = 60
28

THANKS!
Any questions?

Vous aimerez peut-être aussi