Title: Business Statistics Course Notes Rev. 2003
1Business StatisticsCourse Notes Rev. 2003
- Professor Joel M. Calabrese
- San Francisco State University
2Part 1Introduction
3What is Statistics?
- Definitions
- A statistic is a numerical fact or datum
assembled, classified, or tabulated so as to
present significant information about a given
topic. - Statistics The science of preparing and
analyzing such data. - Descriptive Statistics Preparing and presenting
information - Inferential statistics Analyzing Information
and drawing conclusions (inferences).
4Inferential Statistics
- Infer conclusions about a large population using
a small sample from the population Uses
probabilities. - Example TV Ratings Nielsen Media Research.
5TELEVISION RATINGSNielsen Media Research
- NIELSEN PEOPLE METER is programmed with the age
and gender of each household member. Viewers
enter their code when they begin watching
visitors can log their presence as well. The
meter records which channels are tuned by sensing
the frequencies emitted by the cable box, TV or
videocassette recorder. - EVERY DAY, in some 5,000 homes throughout the
U.S., People Meters gather data on who watched
what, when and for how long. - AT STAGGERED TIMES throughout the night, all the
meters call Nielsen's mainframe computer system
in Dunedin, Fla., and transfer their daily
viewing records via modem. - BY MORNING, Nielsen has assembled and processed
its sample of the nation's viewing behavior. TV
executives and other subscribers can log in to
Nielsen's data network to learn which shows were
hits and which flopped. - VERY WEEK subscribers receive a detailed report
chronicling how many Nielsen household viewers
were watching television during any given quarter
hour and how specific programs fared against
their competition.
Source Edgar W. Aust, senior vice president of
engineering and technology for Nielsen Media
Research in Dunedin, Fla.
6Nielsen Data Collection
7Neilsen Media Research
In 1936 engineer Arthur C. Nielsen, Sr., attended
a demonstration at the Massachusetts Institute of
Technology of a mechanical device that could keep
a record of the station to which a radio was
tuned at any given moment. Nielsen bought the
technology practically on the spot and six years
later launched the Nielsen Radio Index, which
analyzed the listening habits of 800 homes.
Later, he adapted the same technology to the new
medium of television, creating a ratings system
that nearly all American broadcasters use today
to help determine the popularity of their
programs. Over the years, Nielsen Media Research
has used several methods to collect viewing
information, including surveys and volunteer
diaries. In 1986 the company supplanted these
with an electronic device called a People Meter.
The meter is now connected to televisions and
telephone lines in about 5,000 households
throughout the U.S. Nielsen households are
selected from a sample that is statistically
representative of the television-viewing
population. Each household receives nominal
compensation--about 50 and occasional gifts--for
their cooperation. In order to keep the sample
representative, viewers can participate for only
two years. As they watch TV, volunteers press
buttons to indicate their presence. The People
Meter records the gender and age of each viewer,
as well as the time spent watching each channel
frequency. Every night the device transmits that
household's data by modem to Nielsen's central
computer in Florida, which assembles the data
into a ratings database. To meet the changing
needs of broadcasters and sponsors, the
technology continues to evolve. In 1986 Nielsen
introduced a system that uses computerized
pattern recognition to identify particular
commercials as they are broadcast. Future
versions of the People Meter now under
development will monitor codes embedded into
digital TV signals to verify which programs are
on the air. They will also use image-recognition
computers to identify viewers the moment they hit
the couch. Source EDGAR W. AUST, senior vice
president of engineering and technology for
Nielsen Media Research in Dunedin, Fla.
8How Accurate Are Statistical Samples?
- Number of U.S. Households in 1999103.9 million.
- Number of Households with TVs100 million.
- Number of Households in Nielsen Sampleabout
5000.Nielsen samples about 1 out of every
20,000 households.
9Outline of Course Topics
- Introduction Data, Variables and Descriptive
Measures (2 weeks) - Basic Probability Theory (2 weeks)
- Probability Distributions (3 weeks)
- Statistical Sampling and Estimation (2 weeks)
- Hypothesis Testing (3 weeks)
- Regression Analysis (3 weeks)
10Definitions
- Subjects Persons or Objects having specific
characteristics we wish to measure. - Variable A measurable characteristic of the
subjects. - Population Set of measurements from all
subjects. - Sample Selected subset of measurements.
11Data and Variables
12Levels of Measurement
- Nominal Data Numbers that label qualitative
differences. - Example Residence Variable
- 1 USA Citizen, but not CA Resident
- 2 USA Citizen California Resident
- 3 Foreign Student
- Ordinal Data Assigned numbers that indicate
rank order. - Example Grade Points
13Levels of Measurement
- Interval Data -- Intervals between numbers can be
compared, but not ratios (no natural zero point). - Examples Calendar Years, Fahrenheit
Temperatures - Ratio Data -- Ratios and Intervals can be
compared (data has a natural zero point). - Examples Height, weight, Length
14Kinds of Variables
- Discrete Variables Values can be represented as
separate, distinct points on a number line. - Example The number of magazines subscribed to by
a student. Possible values 0,1,2,3,... - Continuous Variables Possible values
represented as a continuum on a number line. - Examples Measurements of height, length,
weight, time.
15Subscripts and Summations
- A variable is a list of measurements or
observations. A subscript identifies a
particular observation in the list. - Examples X2, X7, W3
- A summation sign (S) indicates addition.
- Example S X means the sum of all the values
of X
16Rules of Summation
- S cX c S X
- S c nc
- S (X Y) S X S Y
- c a constant
- n total number of observations
- But note S XY does not equal S X S Y
- S X2 does not equal (S X)2
17Applying the Rules
18Using the Rules of Summation
19Calculating the Variance
20Part 2Descriptive Statistics
21Descriptive Statistics
- Measures of Location Where are values located?
- Measures of Variation How spread out are the
data? - Summarizing, Classifying, and Presenting Data
22Measures of Location
- Mean (or Average) S X / n
- Known as for sample.
- Known as m for population.
- Median the middle value, X(n1)/2
- Mode most frequently observed value
- Percentiles shows position of a value
- the pth percentile is the value such that at
least p of all values in the data set are at or
below it and at least (100-p) are at or above
it. - (Note n denotes the sample size N is the
population size).
23Example Ages of playground visitors in
years. Raw Data 2,1,6,2,3,3,7,5,2,4,5,4,6,6,7,
6,3,2,3,6,3,5,6,5,6,2,7,3 S X /
n 120 / 28 4.29 years 1,2,2,2,2,2,3,3,3,3,3,
3,4,4,5,5,5,5,6,6,6,6,6,6,6,7,7,7
Median X(n1)/2 X14.5 4.5 years
Mode 6 years (with secondary mode 3)
24Estimating Percentiles
- Arrange the data in ascending order.
- Compute the subscript of the percentile i
(p/100)n 1/2, where p is the percentile to be
calculated and n is the number of data items. - If i is an integer, then the pth percentile is Xi
- If i is not an integer, then interpolate between
the two X values with subscripts just above and
below i.
25Examples Calculating Percentiles
1,2,2,2,2,2,3,3,3,3,3,3,4,4,5,5,5,5,6,6,6,6,6,6,6,
7,7,7
Estimate the 75th percentile i (p/100)n
½ (75/100)(28) ½ 21.5 75th percentile ?
(X21 X22)/2 (66)/2 6
Estimate the 20th percentile i (p/100)n
(20/100)(28) ½ 6.1 20th percentile ? X6
(0.1)(X7 ? X6) 2 (0.1)(3-2) 2.1
26Other Common Terms
- Quartiles
- The 25th percentile is the first quartile
- The 50th percentile is the second quartile
- The 75th percentile is the third quartile
- The 100th percentile is the fourth quartile
- Deciles
- The 10th percentile is the first decile
- The 20th percentile is the second decile,
- etc.
27Measures of Variation
- Range highest minus lowest value
- Example Range of ages in playground7 years
(oldest) - 1 year (youngest) 6 years - Mean Absolute Deviation (MAD)
The MAD is the average distance from the mean.
28Calculating the MAD
29- Variance
- Known as s2 for sample. Formula is
or
- Known as ?2 for population. Formula is
or
n is the sample size N is the population size
30Calculating the Variance
31Using the Computational Formula for Calculating
the Variance
32The Standard Deviation
- The standard deviation is the square root of the
variance. It is called s for a sample, or ? for
a population. - For the example s ?3.4 1.84
- One use of the standard deviation is the 3-Sigma
Rule. This rule says that it is very unusual to
find any observations in the data greater than
the mean plus 3 times s, and also any
observations less than the mean minus 3 times s.
33Grouping and Presenting Data
- Frequency Distributions
- Absolute Frequencies f(X)
- Relative Frequencies p(X)
- Cumulative Frequencies
- Histograms and Frequency Curves
- Calculating the mean, variance, and standard
deviation with grouped data.
34Definitions
- Absolute Frequency f(X)
- A count of the number of times that a particular
value of the variable X occurs. - Relative Frequency p(X)
- The fraction or percentage of times that a
particular value of X occurs. - Histograms and Frequency Curves
- Graphs of frequencies of X.
35Example Playground Data
Childrens ages 1,2,2,2,2,2,3,3,3,3,3,3,4,4,5,5,5,
5,6,6,6,6,6,6,6,7,7,7
36Graphing f(X)
37Graphing p(X)
38Graphing f(X) Part Two
39Graphing f(X) The Frequency Distribution Curve
40Cumulative Frequencies
Cumulative absolute frequency measures the number
of subjects at or below the indicated value of
X. Cumulative relative frequency measures the
proportion (or percentage) of subjects at or
below the indicated value of X. It also gives an
estimateof the percentile.
41Calculating the Mean of a Frequency Distribution
Using Absolute Frequencies
42Calculating the Mean of a Frequency Distribution
Using Relative Frequencies
43Calculating the Variance and Standard Deviation
- Using absolute frequencies, f(X)
- s2 S (X - )2 . f(X) / (n - 1)
- Using relative frequencies, p(X)
- ?2 S (X - m )2 . p(X) (population formula)
- Note The standard deviation is, as before, the
square root of the variance.
44Example Calculation of Variance
Using relative frequencies Variance ? S (X - m
)2 . p(X)
Note difference between 3.34 and earlier
calculation of 3.4 is due to round-off error and
using population formula.
45Price to Earnings RatiosNatural Resources
Companies
Source Business Week, March 16, 1987
46Frequency Distribution of PE Ratios
47Histogram for PE Ratios
48Relative Frequency Curve for PE Ratios
Example of positively skewed curve.
49Weighted Averages
Where x represents the values of the variable and
w represents the weight on each value.
Formula
Example Calculating GPA
Course Units Grade Grade Points
Comp. Sci 2 C 2 English 5
A 4 Math 3 B 3
50A Note on Transforming Variables
- Suppose you have two variables, x and y, such
that y ax b
a b
VAR(y) a2VAR(x)
STD DEV(y) aSTD DEV(x)
- Example 1 The average wholesale price of a
bottle of wine - at Kermits Restaurant is 6, with a standard
deviation of - 2. The retail price that the customer pays
is equal to the - wholesale price plus a markup of 150 plus a
5 corkage - fee. What are the mean and standard deviation
of the retail - prices?
51Tranforming Variables -- Example 2
- The average daily temperature in June in a
particular location is 68o F with a standard
deviation of 7o F. What is the average and
standard deviation in degrees centigrade? - Note C (5/9)(F - 32)
52Dont Jump to Conclusions!
- Statistics can be misleading. For example,
suppose the two companies below have equivalent
benefit packages. Which one pays more? - Company A
- Mean wage 34,000 a year
- Median wage 40,000 a year
- Modal wage 40,000 a year
- Company B
- Mean wage 30,000 a year
- Median wage 24,000 a year
- Modal wage 24,000 a year
53Look for Reasons Behind Differences
Company A
Employee Number of
Total Classification Employees Payroll
Average
White Collar 600 24 m 40,000 Skilled
Trade 200 6 m 30,000 Unskilled
200 4 m 20,000 Total
1000 34 m 34.000
Company B
Employee Number of
Total Classification Employees Payroll
Average
White Collar 200 8.8 m 44,000 Skilled
Trade 200 6.8 m 34,000 Unskilled
600 14.4 m 24,000 Total
1000 30.0 m 30.000
54Part 3Basic Probability Theory
55Probability
- Definition A probability is a number between 0
and 1, representing the likelihood that a given
event will occur. Examples of events - Coin flip comes up heads
- Roll a 7 throwing dice
- It will rain on Memorial Day
- You have an auto accident next year
- The economy is good next year
- Your new business succeeds
56Calculating Probabilities
- Three Alternatives
- The Classical Approach
- Experience with Long-Run Percentages
- Subjective Judgment
57The Classical Approach
- Works only if the results of a random action or
experiment can be broken down into a collection
of equally likely outcomes. - Collection of all possible outcomes called the
sample space - An event is thought of as an outcome or set of
outcomes (a subset of the sample space)
58Classical Probabilities
- Probability of an event is P(E) F / T
- F Number of outcomes favorable to E
- T Total number of possible outcomes
- Examples of events
- A a coin flip comes up heads
- B at least one head comes up in two coin
flips - C a jack is drawn from a deck of cards
- D a roll of a pair of dice comes up 7
59Outcomes of Rolling Dice
Roll Die 2
60Long-Run Percentages
- Works if the action or experiment of interest is
repeatable under the same general conditions - Probability of an event is P(E) n / N
- N Total number of trials
- n Number of times the event E occurs
- Examples
- Flipping a bottle cap
- Probability of an auto accident
61Subjective Probabilities
- A resort to human judgment -- the degree of
belief of the decision maker. Examples - The economy is good next year
- Next years sales will be high
- A new TV show will be a hit
- Used often in decision analysis, strategic
decision making
62Ad Campaign Example
- A company sent out a questionnaire to 200
consumers to study the effects of a recent
advertising campaign. Results are - 80 people saw the ad (event A)
- 60 people bought the product (event B)
- 40 people saw the ad, but did not purchase the
product - Calculation of probabilities?
63Organizing Ad Campaign Data
Joint Probability Table
64Combining Events
- Intersection of events
- written (A and B) (A Ç B) or (A,B).
- The situation where both events occur. If C
(A Ç B), then C represents the event that both A
and B occur. - Union of events
- written (A or B) (A È B)
- The situation where either event occurs (or
both). If D (A È B), then D is the event that
either A or B or both occur. - Which is greater, P(C) or P(D)?
65Ad Campaign Example (contd)
- Write out the following events in symbols and
find the probabilities - E Saw the ad and also purchased the product
- F Did not see the ad, but did purchase the
product - G Either saw the ad or purchased the product
(or both) - H Did not see the ad did not purchase the
product - J Either did not see the ad, or did not
purchase the product, or both
66Conditional Probability
- Conditional Probability -- The likelihood of an
event, given the occurrence of another event.
Written, for example P(A B) -- probability of
A given that the event B occurs. - Revising the probability of A after finding out
that B has occurred.
67Ad Campaign Example (contd)
- Write out the following in symbols and find the
probabilities - Suppose it is known that someone bought the
product. Now what is the probability that s/he
saw the ad? - If someone has seen the ad, what is the
probability that s/he has also bought the
product? - What is the chance that someone who has not seen
the ad will buy the product? - Among the people who have not bought the product,
what proportion has not seen the ad? - Given that a person saw the ad, what are the
chances that the person did not buy the product?
68Some Definitions
- Two events A and B are mutually exclusive if the
occurrence of one automatically rules out the
occurrence of the other. - A group of events A, B, C, ... are collectively
exhaustive if, taken together, they cover all the
possible outcomes of an action or experiment. - Two events A and B are statistically independent
if the occurrence of one does not affect the
probability that the other occurs this means
that P(A) P(A B) and P(B) P(B A).
69Note on checking for independence
- Check if either
- P(A) P(A B) or P(B) P(B A)
- If either one is true, then A and B are
independent -- no need to check both. - If either is false, then A and B are not
independent
70Multiplication Rule for Probability
The rule P(A Ç B) P(A)? P(B A)
Dividing by P(A), we obtain an alternate formula
Note If A and B are independent, then P(A Ç
B) P(A)? P(B)
71Note on Multiplication Rule
- Switching the roles of A and B, note that (A Ç
B) (B Ç A), thus P(A Ç B) P(B)?
P(A B) - Dividing by P(B), we obtain
72Addition Rule for Probability
- The Rule
- P(A È B) P(A) P(B) - P(A Ç B)
- Note If A and B are mutually exclusive, P(A È
B) P(A) P(B)
73Solving Probability Problems
- There is no general method. Some general
guidelines are - Given a random action or experiment, work out
probabilities by laying out all the possible
outcomes and applying the classical rule. - Given some probabilities, work out others by
applying the multiplication and/or addition
rules. Often joint probability tables or
probability trees are useful for organizing the
information.
74Problem Solving Aids
- Probability Trees.
- Joint Probability Tables.
- Venn Diagrams.
- Formulas for Counting Outcomes
- Multiplication rule for counting
- Formulas for permutations and combinations
75Venn Diagram for Ad Campaign Example
76Probability Tree for Ad Campaign Example
77Flipping the Tree
78Tables vs. Trees
- Given some simple probabilities and joint
probabilities, the information can often be
organized in a joint probability table and used
to derive other probabilities. - A probability tree is often helpful if
- You have some simple probabilities and a series
of independent events or - You have some simple and some conditional
probabilities with a sequence of events.
79Example Exam Problem
A small brewery has two bottling machines.
Machine A produces 40 percent of the bottles,
Machine B produces the rest. On average, one out
of every 20 bottles produced by Machine A is
rejected for some reason, while one out of every
five bottles produced by Machine B is
rejected a. (1 pt) Construct a complete
probability tree for this problem to help answer
the following questions. Label all the branches
and show all the relevant probabilities. b. (1
pt) Suppose a bottle is selected at random. What
is the probability that it will be rejected? c.
(1 pt) Given that a bottle is rejected, what is
the probability that it was produced by Machine B?
80Example Exam Problem
According to recent political polls (S.F.
Chronicle 2/5/98), 67 percent of Americans
approve of President Clintons job performance
(event A) 53 percent think that he had an affair
with Monica Lewinsky (event M). Further analysis
shows that 25 percent think that Clinton had an
affair with Lewinsky and still approve of his job
performance. a. (2 pts) Construct a 2 by 2 table
to help answer the following questions. b. (1 pt)
What proportion of Americans does not think that
President Clinton had an affair with Lewinsky,
yet still does not approve of his job
performance? c. (1 pt) Given that an American
thinks Clinton had an affair with Lewinsky, what
is the probability that s/he approves of
Clintons job performance? d. (1 pt) Find the
probability that an American either does not
approve of Clintons job performance or thinks
that he had an affair with Lewinsky, or both. e.
(1 pt) Is an Americans opinion about the
Lewinsky affair statistically independent from
his/her opinion about Clintons job performance?
(Show your reasoning numerically).
81Example Exam Problem
The director of a large employment agency wishes
to study various characteristics of its job
applicants. In a sample of 8 applicants, 2
applicants have had their current jobs for at
least five years and 3 of the applicants are
college graduates only one of the applicants who
have not graduated from college has had his/her
current job for at least five years. a. (2 ½ pts)
Suppose an applicant from this group is chosen at
random. Find the following probabilities (1) The
applicant is a college graduate. (2) The
applicant is a college graduate who has held
his/her current job for at least five years. (3)
The applicant is either not a college graduate or
someone who has not held his/her current job for
at least five years. (4) That a college graduate
has held his/her job for at least five years. (5)
That someone who has held his/her job for at
least five years is a college graduate. b. (1 pt)
Determine numerically whether being a college
graduate and holding the current job for at least
5 years are statistically independent. c. (1 ½
pt) Do you think that college graduates are more
likely to remain in a job longer? BRIEFLY explain
using appropriate probabilities.
82Example Exam Problem
In drilling for oil in the Arctic, only 10
percent of the holes are "gushers" (very
successful wells), 40 percent are "squirters"
(moderately successful wells), and the rest are
dry holes. Before drilling, however, a geological
test can be done. The test is not perfect the
probability of a favorable test result is 0.85
when a potential site will in fact be a gusher.
The test has a 60 percent chance of a favorable
result when a site will be a squirter and a 90
percent chance of an unfavorable result when the
site will in fact turn out to be a dry hole. a.
Draw a probability tree to represent this
problem. b. What is the probability that the
test will be favorable? c. If the test is
favorable, what is the probability of a
gusher? d. If the test is unfavorable, what is
the probability of a dry hole? e. If the test is
unfavorable, what is the probability of a
squirter?
83Example Exam Problem
- The personnel manager at Megacorp classifies job
applicants as either qualified or unqualified for
the jobs they seek. The manager says that only
25 of the job applicants are qualified. In this
pool of qualified applicants, 20 list high
school as their highest level of education, 50
list college, and the remaining 30 list trade
school. The situation is different among the
unqualified applicants 40 of them list high
school as their highest level of education,
another 40 list trade school, and only 20 list
college. - Draw an appropriate probability tree diagram of
this situation, clearly labeling the event
represented by each branch and its probability. - What is the probability that an applicant is both
qualified and a trade school graduate? - c. What is the unconditional probability that an
applicant comes from a trade school? - d. What is the probability that an applicant is
qualified, given that s/he has a trade school
education?
84Practice Problem
85Practice Problem Independent Events 1
86Practice Problem Independent Events 2