Upload
herbert-gibson
View
224
Download
0
Embed Size (px)
Citation preview
TOPIC 3TOPIC 3
Discrete Probability
Distributions
Discrete Probability
Distributions
Start Thinking Start Thinking
• You’re taking a 33 question multiple choice test. Each question has 4 choices. Clueless on 1 question, you decide to guess. What’s the chance you’ll get it right?
• If you guessed on all 33 questions, what would be your grade? Would you pass?
Discrete Probability DistributionsDiscrete Probability Distributions
Discrete Probability
Distributions
Binomial Poisson Hypergeometric
Negative Binomial
Geometric
Bernoulli Random VariablesBernoulli Random Variables
• Simple random variables with parameter p takes values 0 and 1 with
The values of p :
Then,
The expectation of a Bernoulli random variable
And its variance
10 p pXPandpXP 101
pXE
ppXVar 1
Binomial DistributionsBinomial Distributions
Consider an experiment consisting• n Bernoulli trials (identical trials)• that are independent and• that each have a constant probability p of success (desired
outcome)
Then the total number of successes X is a random variable that has a binomial distribution with parameters n and p, which is written
The probability mass function of a B(n,p) random variable is
for x = 0, 1, 2, …, n with the expectation (mean) and the variance
pnBX ,~
xnx ppx
nxXP
1
pnXE ppnXVar 1
Binomial Distribution ExamplesBinomial Distribution Examples
Number of ‘successes’ in a sample of n observations (trials)
Examples:• Number of reds in 15 spins of roulette wheel
• Number of defective items in a batch of 5 items
• Number correct on a 33 question exam
• Number of customers who purchase out of 100 customers who enter store
ExampleExample
Experiment: Toss 1 coin 5 times in a row. Note number of tails.
What’s the probability of exactly 3 tails ?
What’s the probability of less than or equal to 3? (0, 1,2, or 3) Note: This is known as ‘Cumulative Probability Function’
844.03125.0312.01570.003125.0
)3()2()1()0()3(
3125.05.015.0!3!35
!5)3(
5.0
1!!
!
353
xPxPxPxPxP
xP
p
ppxxn
nqp
x
nxP xnxxnx
Symmetric Binomial DistributionsSymmetric Binomial Distributions
n = 5 p = 0.1 n = 5 p = 0.5
For p = 0.5, written B(n, 0.5), the binomial probability distribution is symmetric. The distribution is symmetric about the expected value of n/2 In above case where B(5, 0.5), the expected value is 2.5.
Proportion of SuccessesProportion of Successes
Let the proportion of successes of a binomial distribution is given by
n
XY
and has an expected value and variance of
n
ppYpYE
12and
As the number of trial n increases, the proportion Y tends to become closer and closer to the success (trial) probability p
0.00
0.25
0.50
0.75
1.00
0 25 50 75 100 125
ExercisesExercises
1) You’re a telemarketer selling service contracts for Macy’s. You’ve sold 20 in your last 100 calls. If you call 12 people tonight, what’s the probability of
a) No sales?
b) Exactly 2 sales?
c) At most 2 sales?
d) At least 2 sales?
2) A multiple choice quiz consists of 10 questions, each with 5 possible answers of which only one is correct. A student passes the quiz if seven or more correct answers obtained. What is the probability that a student who guesses blindly at all of the questions will pass the quiz? What is the probability of passing the quiz if, on each question, a student can eliminate three incorrect answers then guesses between the remaining two?
3) A biologist has a culture consisting of 13 cells. In a period of one hour, independent of the other cells, there is a probability of 0.4 that each of these cells splits into two cells. What is the probability that after one hour the biologist has at least 16 cells? What is the expected number of cells after one hour?
Answers to the ExercisesAnswers to the Exercises
1) p = 20/100 = 0.2 and n = 12
a) No sales means P(X = 0)
b) Exactly 2 sales means P(X = 2)
c) At most 2 sales means P(X ≤ 2)
d) At least 2 sales means P(X ≥ 2)
0687.02.012.00
1210 0120
xnx ppx
nXP
283.02.012.02
122 2122
XP
558.0283.0206.00687.02102 XPXPXPXP
725.0206.00687.0110112322 PPPPPXP
206.02.012.01
121 1121
XP
Answers to the ExercisesAnswers to the Exercises
2) a)
b)
00086.000000010.00000041.0000074.0000786.0
109877
00000010.02.012.010
1010
0000041.02.012.09
109
000074.02.012.08
108
000786.02.012.07
107
10,2.051
101010
9109
8108
7107
PPPPXP
XP
XP
XP
XP
np
172.000098.00098.0044.0117.0
109877
10,5.021
PPPPXP
np
Answers to the ExercisesAnswers to the Exercises
3) a) At least 16 cells means that there are at least three cells split
b)
Jadi paling tidak ada 6 sel yang membelah sehingga ekspektasi jumlah sel setelah 1 jam adalah 13 + 6 = 19
942.0045.0011.00013.01
210113433
045.04.014.02
132
011.04.014.01
131
0013.04.014.00
130
?3,13,4.0
2132
1131
0130
PPPPPPXP
XP
XP
XP
XPnp
2.54.013 pnXE
Poisson DistributionsPoisson Distributions
• A random variable X distributed as a Poisson random variable with parameter λ , which is written
• Has a probability mass function
• λ is the expected (mean) number of ‘successes’
• e = 2.71828 (base of natural logarithm or Euler number)
• Then, the mean or expectation and the variance
PX ~
,2,1,0!
xforx
exXP
x
XVarXE
Poisson Distribution ExamplesPoisson Distribution Examples
• Number of events that occur in an interval
• events per unit time / length / area / volume
• Examples
• Number of customers arriving in 20 minutes
• Number of strikes per year in the U.S.
• Number of defects per lot (group) of DVD’s
Poisson Distribution PropertiesPoisson Distribution Properties
• Constant event probability
• Average of 60/hr is1/min for 60 1-minuteintervals
• One event per interval
• Don’t arrive together
• Independent events
• Arrival of 1 person doesnot affect another’sarrival
Poisson Distribution GraphsPoisson Distribution Graphs
.0
.2
.4
.6
.8
0 1 2 3 4 5
X
P(X)
.0
.1
.2
.3
0 2 4 6 8 10X
P(X)
= 0.5
= 6
ExampleExample
Experiment: Customers arrive at a rate of 72 per hour. What is the probability of 4 customers arriving in 3 minutes?
72 Per Hr. = 1.2 Per Min. = 3.6 Per 3 Min. Interval
1912.0
!4
6.3
!4
6.36.34
e
x
eXP
arrivalsx
ExercisesExercises
1) You work in Quality Assurance for an investment firm. A clerk enters 75 words per minute with 6 errors per hour. What is the probability of 0 errors in a 255-word bond transaction?
2) On average there are about 25 imperfections in 100 meters of optical cable. Use the Poisson distribution to estimate the probability that there are no imperfections in 1 meter of cable. What is the probability that there is no more than one imperfection in 1 meter of cable?
3) Assume that the number of network errors experienced in a day on a local area network (LAN) is distributed as a Poisson random variable. The mean number of network errors experienced in a day is 2.4. What is the probability that in any given daya) Zero network errors will occur?b) Exactly one network error will occur?c) Two or more network errors will occur?d) Less than three network errors will occur?
Answers to the ExercisesAnswers to the Exercises
1) 75 words/min = 75 × 60 = 4500 words/hour.If a 4500-word has 6 errors, then for a 255-word
2) 25 imperfections in 100 meters of cableIn 1 meter cable, there is
712.0
!0
34.0
!0
34.02554500
6
034.0
e
x
eXP
error
x
974.0195.0779.0!1
25.0
!0
25.0101
779.0!0
25.0
!0
25.01100
25
125.0025.0
025.0
eeXPXPXP
e
x
eXP
onimperfecti
x
Answers to the ExercisesAnswers to the Exercises
3)
a)
b)
c)
d)
57.0261.0218.0091.02103
691.0218.0091.011012
218.0!1
4.21
091.0!0
4.2
!0
4.2
14.2
04.2
PPPXP
PPXP
eXP
e
x
eXP
errorx
• For the binomial distribution, the sample data are selected with replacement from a finite population or without replacement from an infinite population.
• For the hypergeometric distribution, the sample data are selected without replacement from a finite population.
• The hypergeometric represents the distribution of the number of items of a certain kind in a random sample of size n drawn without replacement from a population of size N that contains r items of this kind
Hypergeometric DistributionsHypergeometric Distributions
• The hypergeometric distribution has a probability mass function given by
• for max (0, n + r – N) ≤ x ≤ min (n, r) , with• An expected value of
• And a variance of
Hypergeometric DistributionsHypergeometric Distributions
n
Nxn
rN
x
r
xXP
N
nrXE
N
r
N
rn
N
nNXVar 1
1
ExampleExample
• A box of N = 10 computer chips contains r = 5 defective computer chips, and n = 3 chips are selected at random without replacement. What is the probability that the sample contains exactly one defective chip?
4170
238910
245
5
331010
2255
1155
3
10
13
510
1
5
1
.
!!!
!!!
!!!
n
N
xn
rN
x
r
xP
• It is sometimes of interest to count the number of trials performed until the first success occurs.
• The number of trials up to and including the first success in a sequence of independent Bernoulli trials with a constant success probability p has a geometric distribution. The p.m.f. is
• The cumulative distribution function is
• The expected value and the variance are
Geometric DistributionsGeometric Distributions
,3,2,11 1 xforppxXP x
xpxXP 11
2
11
p
pXVarand
pXE
• The number of trials up to and including the r th success in a sequence of independent Bernoulli trials with a constant success probability p has a negative binomial distribution. The p.m.f. is
• for x = r, r + 1, r + 2, r + 3, ….• The expected value and the variance are
Negative Binomial DistributionsNegative Binomial Distributions
rrx ppr
xxXP
11
1
2
1
p
prXVarand
p
rXE
ExampleExample
• When a fisherman catches a fish, it is a young one with a probability of 0.23 and it is returned to the water. On the other hand, an adult fish is kept to be eaten later.
a) What is the expected number of fish caught by the fishermen before an adult fish is caught?
b) What is the probability that the fifth fish caught is the first young fish?
• Suppose that the fisherman wants three fish to eat for lunch
c) What is the probability that the first time the fisherman can stop for lunch is immediately after the sixth fish has been caught?
d) If the fisherman catches five fish, what is the probability that there are sufficient fish for lunch?
Answer to the ExampleAnswer to the Example
a) padult fish = 1 – 0.23 = 0.77
b) pyoung fish = 0.23 , x = 5
c) p = 0.77 , x = 6 , r = 3
d)
29.177.0
11 p
XE
081.023.023.0115 151 ppXP x
056.077.077.0113
161
1
16 336
rrx ppr
xXP
9164.077.077.0113
1577.077.01
13
14
77.077.0113
135435
335334
333
PPPXP
• Consider a sequence of n independent trials where each individual trial can have k outcomes that occur with constant probability values p1, …, pk with p1 + … + pk = 1. The random variables X1, … , Xk that count the number of occurrences of each outcome are said to have a multinomial distribution, and their joint probability mass function is
• satisfying
• The random variables X1, … , Xk expected value and the variance are (they are not independent)
Multinomial DistributionsMultinomial Distributions
kxk
x
kkk pp
xx
nxXxXP
1
11
11 !!
!,,
iiii ppnXVarandpnXE 1
nxx k 1
ExampleExample
• Patients being treated with a particular drug run the risk of being allergic to the drug. A patient’s reaction to the drug is characterized by doctors as being hyper allergic, allergic, mildly allergic, or not allergic, and these have probability values of 0.12, 0.28, 0.33, and 0.27. If there are nine patients,
a) What is the probability that two patients are hyper allergic, one patient is allergic, four patients are mildly allergic, and two patients exhibit no reaction?
b) What is the probability that no patients are hyper allergic, one patient is allergic, four patients are mildly allergic, and four patients exhibit no reaction?
c) What is the expected number of hyper allergic reaction?
d) What is the expected number of allergic reaction?
Answer to the ExampleAnswer to the Example
a)
b)
c)
d)
013.027.033.028.012.0!2!4!1!2
!9
2,4,1,2
2412
4321
XXXXP
011.027.033.028.012.0!4!4!1!0
!9
4,4,1,0
4410
4321
XXXXP
08.112.0911 pnXE
52.228.0922 pnXE