ppt - Department of Mathematics

Report
Chapter 4. Discrete Probability
Distributions
Sections 4.7, 4.8: Poisson and Hypergeometric
Distributions
Jiaping Wang
Department of Mathematical Science
03/04/2013, Monday
The UNIVERSITY of NORTH CAROLINA at CHAPEL HILL
Outline
Poisson: Probability Function
Poisson: Mean and Variance
Hypergeometric: Probability Function
Hypergeometric: Mean and Variance
More Examples
The UNIVERSITY of NORTH CAROLINA at CHAPEL HILL
Part 1. Poisson: Probability
Function
The UNIVERSITY of NORTH CAROLINA at CHAPEL HILL
Consider the probability of the number of accidents that occur at
a particular highway intersection in a period of one week.
Split one week into n subintervals such that
P(One accident in a subinterval)=p
P(No accident in a subinterval) = 1-p
Here we assume p holds for all subintervals and
P(More than two accidents in a subinterval)=0
Also assume n∞, p0 but np=λ.
So, the probability that x accidents happen in these n subintervals is
 λ 
λ −
−
 

The UNIVERSITY of NORTH CAROLINA at CHAPEL HILL
Probability Function
The Poisson probability function:
λ − 
P(X=x)=p(x)=  , x=
!
The distribution function is
F(x)=P(X≤x)=

 λ
=0 !
0, 1, 2, …., for λ> 0
−
Recall that λ denotes the mean number of occurrences in one time period, if there
are t non-overlapped time periods, then the mean would be λt. Poisson distribution
is often referred to as the distribution of rare events.
The UNIVERSITY of NORTH CAROLINA at CHAPEL HILL
Example 4.22
During business hours, the number of calls passing through a
particular relay system averages five per minute.
1. Find the probability that no call will pass through the relay
system during a given minute.
2. Find the probability that no calls will pass through the relay
system during a 2-minute period.
3. Find the probability that three calls will pass through the
relay system during a 2-minute period.
Answer: 1. λ = 5, so P(X=0)=p(0)=50/0! e-5 = e-5 = 0.007.
2. λt=2x5=10, so P(X=0)=p(0)= 100/0! e-10 = e-10 = 0.00005.
3. λt=2x5=10, so P(X=3)=p3)= 103/3! e-10 = 0.0076.
The UNIVERSITY of NORTH CAROLINA at CHAPEL HILL
Example 4.23
Refer to Example 4.22, find the following probabilities:
1. No more than four calls in the given minute.
2. At least four calls in the given minute.
3. Exactly four calls in the given minute.
Answer: λ = 5,
1. so P(X ≤ 4)=F(4)=

5
−5
4
=0.44.
=0 ! 

5
−5
3
=0 ! 
2. P(X ≥ 4) = 1-P(X ≤ 3) = 1 – F(3) = 1 = 1 − 0.265 = 0.735.
3. P(X=4) = P(X ≤ 4) – P(X ≤ 3) = F(4) – F(3) = 0.44 – 0.265 = 0.175.
The UNIVERSITY of NORTH CAROLINA at CHAPEL HILL
Part 2. Poisson:Mean and
Variance
The UNIVERSITY of NORTH CAROLINA at CHAPEL HILL
Mean and Variance




 =  +  +
+ + …
! !
Similarly, we can find V(X)=λ. So
E(X)= V(X) = λ
for Poisson random variable.
The UNIVERSITY of NORTH CAROLINA at CHAPEL HILL
Example 4.24
The manager of an industrial plant is planning to buy a new
machine of either type A or type B. For each day’s operation,
the number of repairs that machine A requires is a Poisson
random variable with mean 0.10t, where t is the time (in hours)
of daily operation. The number of daily repairs Y for machine B
is a Poisson random variable with mean 0.12t. The daily cost of
operating A is CA(t)=20t+40X2; for B, the cost is
CB(t)=16t+40Y2. Assume that the repairs take negligible time
and that each night the machines are to be closed so that they
operate like new machines at the start of each day. Which
machine minimizes the expected daily cost for the following
times of daily operation? 1. 10 hour; 2. 20 hours.
The UNIVERSITY of NORTH CAROLINA at CHAPEL HILL
Answer: E[CA(t)]=20t+40E(X2) = 20t+40[V(X)+E2(X)]
= 24t+0.4t2.
E[CB(t)]=16t+40E(Y2) = 16t+40[V(X)+E2(Y)]
=20.8t+0.576t2.
For Option I: 10 hours. E[CA(t)]= 24(10) + 0.4(10)2 = 280.
E[CB(t)]=20.8(10)+0.576(10)2 = 265.6.
which results in the choice of B.
For Option 2: 20 hours. E[CA(t)]= 24(20) + 0.4(20)2 = 640.
E[CB(t)]=20.8(20)+0.576(20)2 = 646.4.
which results in the choice of A.
The UNIVERSITY of NORTH CAROLINA at CHAPEL HILL
Part 3. Hypergeometric:
Probability Function
The UNIVERSITY of NORTH CAROLINA at CHAPEL HILL
Until now, we learned distributions based on the independent
Bernoulli trials. Now considering a dependent case.
Example: A lot has 10 computer chips, of which 4 are
defective. Now we select two chips randomly without
replacement, what is the probability of choosing one
defective chip?
Answer: P(X=1)=


−

/


The UNIVERSITY of NORTH CAROLINA at CHAPEL HILL
Probability Function
Now we consider a general case: Suppose a lot consists
of N items, of which k are of one type (called
successes) and N-k are of another type (called
failures). Now n items are sampled randomly and
sequentially without replacement. Let X denote the
number of successes among the n sampled items. So
What is P(X=x) for some integer x?
The probability function is:
P(X=x) = p(x) =


−
−


, 0 ≤  ≤  ≤ , 0 ≤  ≤  ≤ 
Which is called hypergeometric probability distribution.
The UNIVERSITY of NORTH CAROLINA at CHAPEL HILL
Example 4.25
Two positions are open in a company. Ten men and five women
have applied for a job at this company and all are equally qualified
for either position. The manager randomly hires two people from
the applicant pool to fill the positions. What is the probability that
a man and a woman were chosen?
Answer: N=15, k=10 men (as successes), x=1, n=2. So using hypergeometric
probability function to have
P(X=1) = p(1) =
10
1
15−10
1
15
2
10
= 21
The UNIVERSITY of NORTH CAROLINA at CHAPEL HILL
Part 4. Hypergeometric:Mean
and Variance
The UNIVERSITY of NORTH CAROLINA at CHAPEL HILL
Mean and Variance

  =


  =


−

−
−
The UNIVERSITY of NORTH CAROLINA at CHAPEL HILL
Example 4.26
In an assembly line production of industrial robots, gearbox
assemblies can be installed in 1 minute each, if the holes have
been properly drilled in the boxes, and in 10 minutes each if the
holes must be installed. There are 20 gearboxes in stock, and 2 of
them have been improperly drilled holes. From the 20 gearboxes
available, 5 are selected randomly fro installation in the next 5
robots in line.
1. Find the probability that all 5 gearboxes will fit properly.
2. Find the expected value, the variance and the standard deviation of the time
it will take to install these five gearboxes.
The UNIVERSITY of NORTH CAROLINA at CHAPEL HILL
Answer: N=20, k=2 (as successes for noncomforming boxes), n=5
1. x=0 means all of the 5 boxes fit properly, so P(X=0)=p(0)=
2
0
20−2
5
20
5
= 0.55
2. Total time T=10X+(5-X) = 9X+5, and
E(X)=n(k/N)=5(2/20)=0.5,
V(X)=n(k/N)(1-k/N)[(N-n)/(N-1)]=5(2/20)(1-2/20)[(20-5)/(20-1)]=0.355
So E(T)=9E(X)+5=9.5, V(T)=(9)2V(X)=28.755,
and the standard deviation is [V(T)]1/2=(28.755)1/2=5.4(minutes).
The UNIVERSITY of NORTH CAROLINA at CHAPEL HILL
Part 5. More Examples
The UNIVERSITY of NORTH CAROLINA at CHAPEL HILL
Additional Example 1
A certain kind of sheet metal has, on the average, five defects per
10 square feet. If we assume a Poisson distribution, what is the
probability that a 15-square feet sheet of the metal will have at
least six defects?
Answer:λs=5(15/10)=7.5,
so P(X≥6)=1-P(X≤5)=1-
7.5
5
=0 !

.
 − 75
The UNIVERSITY of NORTH CAROLINA at CHAPEL HILL
Additional Example 2
Let X be a Poisson random variable with mean λ.
If P(X = 1|x≤ 1) = 0.8, what is the value of ?
Answer:
P(X=1|X ≤ 1) = P(X=1, X ≤ 1)/P(X≤1)=P(X=1)/P(X≤1)= λ/(λ+1)=0.8
λ=4.
The UNIVERSITY of NORTH CAROLINA at CHAPEL HILL
Additional Example 3
A box contains 10 white and 15 black marbles. Let X denote the
number of white marbles in a selection of 10 marbles selected at
random and without replacement. Find Var(X)/E(X).
Answer:
N=25, k=10, n=10, E(X)=n(k/N)=10(10/25)=4, V(X)=n(k/N)(1-k/N)[(Nk)/(N-1)]=4(1-10/25)(15/24)=1.5, so V(X)/E(X)=0.375
The UNIVERSITY of NORTH CAROLINA at CHAPEL HILL

similar documents