1 1 slide 2009 university of minnesota-duluth, econ-2030 (dr. tadesse) chapter 5: probability...
Post on 19-Dec-2015
222 views
TRANSCRIPT
1 1 Slide
Slide
2009 University of Minnesota-Duluth, Econ-2030 (Dr. Tadesse)
Chapter 5: Probability Distributions:
Discrete Probability Distributions
2 2 Slide
Slide
2009 University of Minnesota-Duluth, Econ-2030 (Dr. Tadesse)
Learning Objectives
Identifying Types of Discrete Probability Distribution and their Respective Functional Representations
Calculating the Mean and Variance of a Discrete Random Variable with Each of the Different Discrete Probability Distribution.
3 3 Slide
Slide
2009 University of Minnesota-Duluth, Econ-2030 (Dr. Tadesse)
Random Variable
4 4 Slide
Slide
2009 University of Minnesota-Duluth, Econ-2030 (Dr. Tadesse)
Random Variables
Any variable that is used to represent the outcomes any experiment of interest to us is called Random Variable.
A random variable can assume (take) any value (Positive, negative, zero; finite, infinite; continuous and discrete values).
Depending upon the values they take, we can identify two types of random variables:
1. Discrete Variables2. Continuous Variables.
5 5 Slide
Slide
2009 University of Minnesota-Duluth, Econ-2030 (Dr. Tadesse)
Both types of variables can assume either a finite number of values or an infinite sequence of values. Both types of variables can assume either a finite number of values or an infinite sequence of values.
6 6 Slide
Slide
2009 University of Minnesota-Duluth, Econ-2030 (Dr. Tadesse)
Let x = number of TVs sold at a given store in one day. The number of TV units that can be sold in a given day is finite. It is also discrete: (0, 1, 2, 3, 4).
X can be considered as Discrete Random Variable
Let x = number of TVs sold at a given store in one day. The number of TV units that can be sold in a given day is finite. It is also discrete: (0, 1, 2, 3, 4).
X can be considered as Discrete Random Variable
Example: JSL Appliances
Discrete random variable with a finite number of values
7 7 Slide
Slide
2009 University of Minnesota-Duluth, Econ-2030 (Dr. Tadesse)
Let Y = number of customers arriving in a store in one day. Y can take on the values 0, 1, 2, . . .
Let Y = number of customers arriving in a store in one day. Y can take on the values 0, 1, 2, . . .
Example: JSL Appliances
Discrete random variable with an infinite sequence of values
We can count the customers arriving. However, there is nofinite upper limit on the number that might arrive.
8 8 Slide
Slide
2009 University of Minnesota-Duluth, Econ-2030 (Dr. Tadesse)
Random Variables
Question Random Variable x Type
Familysize
X = Number of dependents reported on tax return
Discrete
Distance fromhome to store
Y = Distance in miles from home to the store site
Continuous
Own dogor cat
Z = 1 if own no pet; = 2 if own dog(s) only; = 3 if own cat(s) only; = 4 if own dog(s) and cat(s)
Discrete
9 9 Slide
Slide
2009 University of Minnesota-Duluth, Econ-2030 (Dr. Tadesse)
Probability Distributions
10 10 Slide
Slide
2009 University of Minnesota-Duluth, Econ-2030 (Dr. Tadesse)
The probability distribution for a random variable is a distributionthat describes the values that the random variable of interest takes. The probability distribution for a random variable is a distributionthat describes the values that the random variable of interest takes.
Probability Distributions
The probability distribution is defined by a probability function, denoted by f(x)---the probability of the values of the random variable.
The probability distribution is defined by a probability function, denoted by f(x)---the probability of the values of the random variable.
11 11 Slide
Slide
2009 University of Minnesota-Duluth, Econ-2030 (Dr. Tadesse)
For any probability function the following conditions must besatisfied:For any probability function the following conditions must besatisfied:
Probability Distributions
1. f(x) > 01. f(x) > 0
2. f(x) = 12. f(x) = 1
12 12 Slide
Slide
2009 University of Minnesota-Duluth, Econ-2030 (Dr. Tadesse)
Discrete Probability Distributions
13 13 Slide
Slide
2009 University of Minnesota-Duluth, Econ-2030 (Dr. Tadesse)
A Discrete Probability Distribution is a tabular, graphic orFunctional representation of a Random Variable with discrete outcomes that follows the principle of probability distribution
A Discrete Probability Distribution is a tabular, graphic orFunctional representation of a Random Variable with discrete outcomes that follows the principle of probability distribution
Discrete Probability Distributions
f(x) > 0f(x) > 0
f(x) = 1f(x) = 1
14 14 Slide
Slide
2009 University of Minnesota-Duluth, Econ-2030 (Dr. Tadesse)
• Using past data on sales, a tabular representation of the probability distribution for TV sales was developed.
Number Units Sold of Days
0 80 1 50 2 40 3 10 4 20
200
x f(x) 0 .40 1 .25 2 .20 3 .05 4 .10 1.00
80/200
Discrete Probability Distributions--Example
16 16 Slide
Slide
2009 University of Minnesota-Duluth, Econ-2030 (Dr. Tadesse)
1. Uniform Distribution
2. Binomial Distribution
3. Poisson Distribution
A Discrete Probability Distribution can assume one or more of the Following Distributions:
4. Hyper-Geometric Distribution
17 17 Slide
Slide
2009 University of Minnesota-Duluth, Econ-2030 (Dr. Tadesse)
5.1) Discrete Uniform Probability Distribution
The probability distribution of a discrete probability distribution is given by the following formula. The probability distribution of a discrete probability distribution is given by the following formula.
f(x) = 1/nf(x) = 1/n
where:n = the number of values the random variable may assume
the values of the
random variable
are equally likely
18 18 Slide
Slide
2009 University of Minnesota-Duluth, Econ-2030 (Dr. Tadesse)
where: f(x) = the probability of x successes in n trials n = the number of trials p = the probability of success on any one trial
( )!( ) (1 )
!( )!x n xn
f x p px n x
5.2) Binomial Distribution
The Probability distribution of a Binomial Distribution is given by the following function
19 19 Slide
Slide
2009 University of Minnesota-Duluth, Econ-2030 (Dr. Tadesse)
The Probability of a Poisson Distribution is given by the following function
5.3) Poisson Distribution
f xex
x( )
!
where:f(x) = probability of x occurrences in an interval = mean number of occurrences in an interval e = 2.71828
20 20 Slide
Slide
2009 University of Minnesota-Duluth, Econ-2030 (Dr. Tadesse)
The Probability of a Hyper-geometric Distribution is given by the following Function
5.4) Hypergeometric Distribution
n
N
xn
rN
x
r
xf )( for 0 < x < r
where: f(x) = probability of x successes in n trials n = number of trials N = number of elements in the population r = number of elements in the population
labeled success
21 21 Slide
Slide
2009 University of Minnesota-Duluth, Econ-2030 (Dr. Tadesse)
5.1) Discrete Uniform Probability Distribution
The probability distribution of a discrete probability distribution is given by the following formula. The probability distribution of a discrete probability distribution is given by the following formula.
f(x) = 1/nf(x) = 1/n
where:n = the number of values the random variable may assume
the values of the
random variable
are equally likely
22 22 Slide
Slide
2009 University of Minnesota-Duluth, Econ-2030 (Dr. Tadesse)
Expected Value (Mean) and Variance of Discrete Probability Distributions
23 23 Slide
Slide
2009 University of Minnesota-Duluth, Econ-2030 (Dr. Tadesse)
Expected Value (Mean) for ……..Discrete Uniform Distribution
The expected value, or mean, of a random variable is a measure of its central location. The expected value, or mean, of a random variable is a measure of its central location.
E(x) = = xf(x)E(x) = = xf(x)
24 24 Slide
Slide
2009 University of Minnesota-Duluth, Econ-2030 (Dr. Tadesse)
Variance and Standard Deviation of Discrete Uniform Probability Distribution
The variance summarizes the variability in the values of a random variable. The variance summarizes the variability in the values of a random variable.
The standard deviation, , is defined as the square root of the variance. The standard deviation, , is defined as the square root of the variance.
Var(x) = 2 = (x - )2f(x)Var(x) = 2 = (x - )2f(x)
25 25 Slide
Slide
2009 University of Minnesota-Duluth, Econ-2030 (Dr. Tadesse)
• The likelihood of selling TV sets in any given day is considered equally likely (Uniform). The following table summarizes sales data on the past 200 days.
Number Units Sold of Days
0 80 1 50 2 40 3 10 4 20
200
x f(x) 0 .40 1 .25 2 .20 3 .05 4 .10
Example-TV Sales in a Given Store
Given this data what is the Average Number of TVs sold in a day?
x F(x) 0 .40 1 .65 2 .85 3 .90 4 1.00
26 26 Slide
Slide
2009 University of Minnesota-Duluth, Econ-2030 (Dr. Tadesse)
Given the data, we can find the Expected Value (Mean Number) of TVs sold in a day as follows
expected number of TVs sold in a day
x f(x) xf(x) 0 .40 .00 1 .25 .25 2 .20 .40 3 .05 .15 4 .10 .40
E(x) = 1.20
Expected Value (MEAN) and Variance
E(x) = = xf(x)E(x) = = xf(x)
27 27 Slide
Slide
2009 University of Minnesota-Duluth, Econ-2030 (Dr. Tadesse)
01234
-1.2-0.2 0.8 1.8 2.8
1.440.040.643.247.84
.40
.25
.20
.05
.10
.576
.010
.128
.162
.784
x - (x - )2 f(x) (x - )2f(x)x
TVs squaredStandard deviation of daily sales = 1.2884 TVs
Find the Variance and Standard Deviation of the Number of TV Sold in a given day.
Var(x) = 2 = (x - )2f(x)= 1.66Var(x) = 2 = (x - )2f(x)= 1.66
Var(x) = 2 = (x - )2f(x)Var(x) = 2 = (x - )2f(x)
28 28 Slide
Slide
2009 University of Minnesota-Duluth, Econ-2030 (Dr. Tadesse)
5.2) The Binomial Distribution
29 29 Slide
Slide
2009 University of Minnesota-Duluth, Econ-2030 (Dr. Tadesse)
5.2) Binomial Distribution
Four Properties of a Binomial Experiment
3. The probability of a success, denoted by p, does not change from trial to trial.3. The probability of a success, denoted by p, does not change from trial to trial.
4. The trials are independent.4. The trials are independent.
2. Only two outcomes, success and failure, are possible on each trial.2. Only two outcomes, success and failure, are possible on each trial.
1. The experiment consists of a sequence of n identical trials.1. The experiment consists of a sequence of n identical trials.
30 30 Slide
Slide
2009 University of Minnesota-Duluth, Econ-2030 (Dr. Tadesse)
5.2) The Binomial Distribution
Typical Examples of a Binomial Experiment:
• Lottery: Win or Lose
• Election: A Candidate Wins or Loses• Gender of an Employee: is Male or Female• Flipping a coin: Heads or Tails
31 31 Slide
Slide
2009 University of Minnesota-Duluth, Econ-2030 (Dr. Tadesse)
5.2) The Binomial Distribution
Our interest is in the number of successes occurring in the n trials. Our interest is in the number of successes occurring in the n trials.
let X denote the number of successes occurring in the n trials.let X denote the number of successes occurring in the n trials.
32 32 Slide
Slide
2009 University of Minnesota-Duluth, Econ-2030 (Dr. Tadesse)
where:n = the number of trials p = the probability of success on any one trial f(x) = the probability of x successes in n trials
( )!( ) (1 )
!( )!x n xn
f x p px n x
5.2) The Binomial Distribution
Binomial Probability Function
33 33 Slide
Slide
2009 University of Minnesota-Duluth, Econ-2030 (Dr. Tadesse)
( )!( ) (1 )
!( )!x n xn
f x p px n x
5.2) The Binomial Distribution
!!( )!
nx n x
( )(1 )x n xp p
Binomial Probability Function
Probability of a particular sequence of outcomes
with x successes
Probability of a particular sequence of outcomes
with x successes
Number of experimental outcomes providing exactly
x successes in n trials
Number of experimental outcomes providing exactly
x successes in n trials
34 34 Slide
Slide
2009 University of Minnesota-Duluth, Econ-2030 (Dr. Tadesse)
Binomial Distribution
Example: Evans ElectronicsA local Electronics company is concerned about it low retention of employees. In recent years, management has seen an annual turnover of 10% in its hourly employees. Thus, for any hourly employee chosen at random, the company estimates that there is 0.1 probability that the person will leave the company in a year time.
35 35 Slide
Slide
2009 University of Minnesota-Duluth, Econ-2030 (Dr. Tadesse)
Binomial Distribution
Given the above information, if we randomly select 3 hourly employees, what is the probability that 1 of them will leave the company in one year ?
f xn
x n xp px n x( )
!!( )!
( )( )
1
1 23!(1) (0.1) (0.9) 3(.1)(.81) .243
1!(3 1)!f
Let: p = .10, n = 3, x = 1Solution:
37 37 Slide
Slide
2009 University of Minnesota-Duluth, Econ-2030 (Dr. Tadesse)
Using Tables of Binomial Probabilities
n x .05 .10 .15 .20 .25 .30 .35 .40 .45 .50
3 0 .8574 .7290 .6141 .5120 .4219 .3430 .2746 .2160 .1664 .12501 .1354 .2430 .3251 .3840 .4219 .4410 .4436 .4320 .4084 .37502 .0071 .0270 .0574 .0960 .1406 .1890 .2389 .2880 .3341 .37503 .0001 .0010 .0034 .0080 .0156 .0270 .0429 .0640 .0911 .1250
p
Binomial Distribution
Page….383-388
38 38 Slide
Slide
2009 University of Minnesota-Duluth, Econ-2030 (Dr. Tadesse)
Mean and Variance of A Binomial Distribution
(1 )np p
E(x) = = np
Var(x) = 2 = np(1 - p)
Expected Value
Variance
Standard Deviation
39 39 Slide
Slide
2009 University of Minnesota-Duluth, Econ-2030 (Dr. Tadesse)
3(.1)(.9) .52 employees
E(x) = = 3(.1) = .3 employees out of 3
Var(x) = 2 = 3(.1)(.9) = .27
Expected Value
Variance
Standard Deviation
Given that p=0.1, for the 3 randomly selected hourly employees, what is the expected number and variance of workers who might leave the company this year?
40 40 Slide
Slide
2009 University of Minnesota-Duluth, Econ-2030 (Dr. Tadesse)
5.3) Poisson Distribution
41 41 Slide
Slide
2009 University of Minnesota-Duluth, Econ-2030 (Dr. Tadesse)
Poisson distribution refers to the probability distribution of a trial that involves cases of rare events that occur over a fixed time interval or within a specified region
6.4. Poisson Distribution
42 42 Slide
Slide
2009 University of Minnesota-Duluth, Econ-2030 (Dr. Tadesse)
Examples….
• The number of errors a typist makes per page• The number of cars entering a service station per hour• The number of telephone calls received by a switchboard per
hour.
• The number of Bank Failures During a given Economic Recession.
• The number of housing foreclosures in a given city during a given year.
• The number of car accidents in one day on I-35 stretch from the Twin Cities to Duluth
6.4. Poisson Distribution
43 43 Slide
Slide
2009 University of Minnesota-Duluth, Econ-2030 (Dr. Tadesse)
We use a Poisson distribution to estimate the number of occurrences of such discrete incidents over a specified interval of time or space
We use a Poisson distribution to estimate the number of occurrences of such discrete incidents over a specified interval of time or space
Thus a Poisson distributed random variable is discrete; Often times it assumes an infinite sequence of values (x = 0, 1, 2, . . . ).
Thus a Poisson distributed random variable is discrete; Often times it assumes an infinite sequence of values (x = 0, 1, 2, . . . ).
Poisson Distribution
44 44 Slide
Slide
2009 University of Minnesota-Duluth, Econ-2030 (Dr. Tadesse)
The number of successes (events) that occur in a certain time interval is independent of the number of successes that occur in another time interval.
The probability of a success in a certain time interval is• the same for all time intervals of the same size,• proportional to the length of the interval.
The probability that two or more successes will occur in an interval approaches zero as the interval becomes smaller.
Properties of a Poisson Experiment
45 45 Slide
Slide
2009 University of Minnesota-Duluth, Econ-2030 (Dr. Tadesse)
Poisson Probability Function
Poisson Distribution
f xex
x( )
!
where:f(x) = probability of x occurrences in an interval = mean number of occurrences in an interval e = 2.71828
46 46 Slide
Slide
2009 University of Minnesota-Duluth, Econ-2030 (Dr. Tadesse)
On Average 6 Patients per hour arrive at the emergency room of Mercy Hospital on
weekend evenings.
What is the probability of 4 arrivals in30 minutes on a weekend evening?
Poisson Distribution--Example
MERCY
47 47 Slide
Slide
2009 University of Minnesota-Duluth, Econ-2030 (Dr. Tadesse)
Poisson Distribution-Example
Using the Poisson Probability Function
4 33 (2.71828)(4) .1680
4!f
MERCY
m= 6/hour = 3/half-hour, P(x = 4)?
f xex
x( )
!
48 48 Slide
Slide
2009 University of Minnesota-Duluth, Econ-2030 (Dr. Tadesse)
Poisson Distribution
Using Poisson Probability Tables
x 2.1 2.2 2.3 2.4 2.5 2.6 2.7 2.8 2.9 3.00 .1225 .1108 .1003 .0907 .0821 .0743 .0672 .0608 .0550 .04981 .2572 .2438 .2306 .2177 .2052 .1931 .1815 .1703 .1596 .14942 .2700 .2681 .2652 .2613 .2565 .2510 .2450 .2384 .2314 .22403 .1890 .1966 .2033 .2090 .2138 .2176 .2205 .2225 .2237 .22404 .0992 .1082 .1169 .1254 .1336 .1414 .1488 .1557 .1622 .16805 .0417 .0476 .0538 .0602 ..0668 .0735 .0804 .0872 .0940 .10086 .0146 .0174 .0206 .0241 .0278 .0319 .0362 .0407 .0455 .05047 .0044 .0055 .0068 .0083 .0099 .0118 .0139 .0163 .0188 .02168 .0011 .0015 .0019 .0025 .0031 .0038 .0047 .0057 .0068 .0081
MERCY
Page---390-395
50 50 Slide
Slide
2009 University of Minnesota-Duluth, Econ-2030 (Dr. Tadesse)
Mean and Variance of Poisson Distribution
Another special property of the Poisson distribution is that the mean and variance are equal.
Another special property of the Poisson distribution is that the mean and variance are equal.
m = s 2
51 51 Slide
Slide
2009 University of Minnesota-Duluth, Econ-2030 (Dr. Tadesse)
Poisson DistributionMERCY
Variance for Number of ArrivalsDuring 30-Minute Periods
m = s 2 = 3 m = s 2 = 3
52 52 Slide
Slide
2009 University of Minnesota-Duluth, Econ-2030 (Dr. Tadesse)
5.4) Hyper-Geometric Distribution
53 53 Slide
Slide
2009 University of Minnesota-Duluth, Econ-2030 (Dr. Tadesse)
Hyper-geometric Distribution
The hyper-geometric distribution is closely related to the binomial distribution. The hyper-geometric distribution is closely related to the binomial distribution.
However, for the hyper-geometric distribution: However, for the hyper-geometric distribution:
the trials are not independent, and the trials are not independent, and
the probability of success changes from trial to trial. the probability of success changes from trial to trial.
54 54 Slide
Slide
2009 University of Minnesota-Duluth, Econ-2030 (Dr. Tadesse)
Hyper-geometric Probability Function
Hyper-geometric Distribution
n
N
xn
rN
x
r
xf )( for 0 < x < r
where: f(x) = probability of x successes in n trials n = number of trials N = number of elements in the population r = number of elements labeled as success
55 55 Slide
Slide
2009 University of Minnesota-Duluth, Econ-2030 (Dr. Tadesse)
Hyper-geometric Probability Function
Hyper-geometric Distribution
( )
r N r
x n xf x
N
n
for 0 < x < r
number of waysn – x failures can be selectedfrom a total of N – r failures
in the populationnumber of ways
x successes can be selectedfrom a total of r successes
in the populationnumber of ways
a sample of size n can be selectedfrom a population of size N
56 56 Slide
Slide
2009 University of Minnesota-Duluth, Econ-2030 (Dr. Tadesse)
Hyper-geometric Distribution
( )r
E x nN
2( ) 11
r r N nVar x n
N N N
Mean
Variance
57 57 Slide
Slide
2009 University of Minnesota-Duluth, Econ-2030 (Dr. Tadesse)
Hyper-geometric Distribution: Inspection
Electric fuses produced by a given company are packed in boxes. Each box carries 12 units of electric fuses. The role of the inspector in the company that manufactures the fuses is to make sure that all fuses in each box are in good condition.
Consider the following scenario: A worker inadvertently places 5 defective items in a box. As part of her job, the inspector randomly selects 3 fuses from the box that contains the defective fuses.
58 58 Slide
Slide
2009 University of Minnesota-Duluth, Econ-2030 (Dr. Tadesse)
Hyper-geometric Distribution: Inspection
1. What is the probability that none of the three randomly selected fuses are defective?
2. What is the probability that the inspector finds only one of the three randomly selected fuses to be defective?
3. What is the probability that the inspector finds at least one of the three fuses defective?
59 59 Slide
Slide
2009 University of Minnesota-Duluth, Econ-2030 (Dr. Tadesse)
Hyper-geometric Distribution: Inspection
1. What is the probability that none of the three randomly selected fuses are defective?
Solution:N=12; r=5; n=3; x=0;
P(x=0)?
1591.0220
35
!9!3!12!4!3!7
!5!0!5
)(
))(()0(
123
51203
50
xf
60 60 Slide
Slide
2009 University of Minnesota-Duluth, Econ-2030 (Dr. Tadesse)
Hyper-geometric Distribution: Inspection
2. What is the probability that the inspector finds only one of the three randomly selected fuses to be defective?
Solution:N=12; r=5; n=3; x=1;
P(x=1)?
4773.0220
215
!9!3!12!5!2!7
!4!1!5
)(
))(()1(
123
51213
51
X
xf
61 61 Slide
Slide
2009 University of Minnesota-Duluth, Econ-2030 (Dr. Tadesse)
Hyper-geometric Distribution: Inspection
2. What is the probability that the inspector finds at least one of the three fuses defective?
Solution:N=12; r=5; n=3; x 1
8409.0
1591.01
)0(1
?)1(
Xp
xf
62 62 Slide
Slide
2009 University of Minnesota-Duluth, Econ-2030 (Dr. Tadesse)
Hyper-geometric Distribution
Example: NevereadyBob has removed two dead batteries from a flashlight and inadvertently mingled them with the two good batteries that he intended to use as replacements. The four batteries look identical.
Then Bob randomly selects two of the four batteries. What is the probability that he selects the two good batteries?
ZAP
ZA
P
ZAPZAP
63 63 Slide
Slide
2009 University of Minnesota-Duluth, Econ-2030 (Dr. Tadesse)
Hyper-geometric Distribution
Using the Hyper-geometric Function
2 2 2! 2!
2 0 2!0! 0!2! 1( ) .167
4 4! 62 2!2!
r N r
x n xf x
N
n
N = 4 = number of batteries in total r = 2 = number of good batteries in total
x = 2 =number of good batteries to be selected n = 2 = number of batteries selected
167.06
1
!2!2
!4!2!0
!2
!0!2
!2
66 66 Slide
Slide
2009 University of Minnesota-Duluth, Econ-2030 (Dr. Tadesse)
Hyper-geometric Distribution
When the population size is large, a hyper-geometric distribution can be approximated by a binomial distribution with n trials and a probability of success p = (r/N).
When the population size is large, a hyper-geometric distribution can be approximated by a binomial distribution with n trials and a probability of success p = (r/N).
67 67 Slide
Slide
2009 University of Minnesota-Duluth, Econ-2030 (Dr. Tadesse)
Hyper-geometric Distribution
Consider a hyper-geometric distribution with n trials and let p = (r/n) denote the probability of a success on the first trial.
If the population size is large, the term (N – n)/(N – 1) approaches 1.
The expected value and variance can be written E(x) = np and Var(x) = np(1 – p).
Note that these are the expressions for the expected value and variance of a binomial distribution.