BUSINESS STATISTICS

CHAPTER 1: Data and statistics

What is Statistics? Elements, Variables, and Observations

• The term statistics can refer to numerical facts such as • Elements

averages, medians, percentages, and maximums that - are the entities on which data are collected.
help us understand a variety of business and economic • Variable
situations. - is a characteristic of interest for the
• Statistics can also refer to the art and science of elements.
collecting, analyzing, presenting, and interpreting • Observations
data. - The set of measurements obtained for a
particular element.
Applications in Business and Economics
• A data set with n elements contains n
• Accounting obser vations.
- Public accounting firms use statistical sampling • The total number of data values in a complete
procedures when conducting audits for their clients. data set is the number of elements multiplied by
• Economics the number of variables.
- Economists use statistical information in making
forecasts about the future of the economy or some Scales of Measurement
aspect of it.
• Finance • Scales of measurement include
- Financial advisors use price-earnings ratios and • Nominal
dividend yields to guide their investment advice. • Ordinal
• Marketing • Inter val
- Electronic point-of-sale scanners at retail • Ratio
checkout counters are used to collect data for a
variety of marketing research applications. • The scale determines the amount of information
• Production contained in the data.
- A variety of statistical quality control charts are • The scale indicates the data summarization and
used to monitor the output of a production process. statistical analyses that are most appropriate.
• Information Systems
- A variety of statistical information helps NOMINAL
administrators assess the performance of computer • Data are labels or names used to identify an
net works. attribute of the element.
• A nonnumeric label or numeric code may be used.
Data and Data Sets
ORDINAL
• Data are the facts and figures collected, analyzed, and • The data have the properties of nominal data
summarized for presentation and interpretation. and the order or rank of the data is meaningful.
• All the data collected in a particular study are referred • A nonnumeric label or numeric code may be used.
to as the data set for the study.
INTERVAL CROSS-SECTIONAL DATA
·

• Data are collected at the same or

• The data have the properties of ordinal data, approximately the same point in time.
and the inter val bet ween observations is • Useful in Time Series Data
expressed in terms of a fixed unit of measure.
Example:
• Interval data are always numeric. Data detailing different varlables like status, Per
capita GDP, Fitch rating for 60 different WTO
nations at the same point in time.
RATIO
• The data have all the properties of interval
data and the ratio of t wo values is TIME SERIES
• Data are collected over several time periods.
meaningful.
• Variables such as distance, height, weight, and Example:
time use the ratio scale. U.S average price per gallon of conventional
regular gasoline bet ween 2010 and 2015
• This scale must contain a zero value that
indicates that nothing exists for the variable Graphs of time series help analysts understand
at the zero point. • what happened in the past,
• identify any trends over time, and
• project future values for the time series
Categorical and Quantitative Data
DATA SOURCES
• Data can be further classified as being
categorical or quantitative. Existing Sources
• Internal company secords - almost any
• The statistical analysis that is appropriate department
depends on whether the data for the variable • Business database ser vices - Dow Jones & Co.
is categorical or quantitative. • Government agencies - U.S. Department of
labor
• Industry associations - Travel Indusry
• In general, there are more alternatives for Association of America
statistical analysis when the data are • Special interest organizations - Graduate
Management Association Council (GMAT)
quantitative. • Internet - more and more firms

Categorical Data
Statistical Studies - Observational
• In observational (nonexperimental) eludies no
• Labels or names used to identify an attribute attempt Is made to control or influence the
variables of Interest
of each element
• Often referred to as qualitative data - A sur vey is a good example
• Use either the nominal or ordinal scale of • Studies of smokers and nonsmokers are
observational studles because researchers do
measurement not determine or control who will smoke and
• Can be either numeric or nonnumeric who will not smoke
• Appropriate statistical analysis is rather
Statistical Studies - Experimental
limited • In experimental studies the variable of
interest is first identified.
Quantitative Data • Then one or more variables are identifled and
controlled so that data can be obtained about
how they influence the variable of interest.
• Quantitative data indicate how many or how - The largest experimental study ever
much: conducted is believed to be the 1954 Public Health
Service experiment for the Salk polio vaccine.
- discrete, if measuring how many Nearly t wo million U.S. children (grades 1- 3) were
- continuous, if measuring how much selected.
• Quantitative data are always numeric
• Ordinary arithmetic operations are
meaningful for quantitative data.
 Data Acquisition Considerations Statistical Analysis Using Microsoft Excel

Time Requirement • Statisticians often use computer soft ware to

• Searching for information can be time consumine perform the statistical computations required
• information may no longer be useful by the time in with large amounts of data.
is available • Many of the data sets in this book are
available on the website that accompanies the
Cost of Acquisition book.
• Organizations often charge for information ten • The data sets are in Microsoft Excel format.
when it is not their primary business activity • Also, the Excel add-in StatTools can be
downloaded from the website.
Data Errors
• Using any data that happen to be available or
were acquired with little care can test to misleading Analytics
Information
Scientific process of transforming data into
insight for making better decisions.
Descriptive Statistics
Types
• Most of the statistical information in
newspapers, magazines, company reports, and Descriptive analysis
other publications consists of data that are • Analytical techniques that describe what
summarized and presented in a form that is easy happened in the past.
to understand.
• Such summaries of data, which may be tabular, Predictive analysis
graphical, or numerical, are referred to as • Analytical techniques that use models
descriptive statistics. constructed from past data te predict future.
• Helps assess the impact the impact of one
Example: Hudson Auto Repair varlable on another

The manager of Hudson Auto wauld like to have a Prescriptive analysis

better understanding of the cost of parts used in • Analytical techniques that yield a best
the engine tune-ups performed in her shop. She course of action to take.
examines 50 customer invoices for tune-ups. The
costs of parts, rounded to the nearest dollar, are
listed on the next slide. Data Warehousing

• Organizations obtain large amounts of data

Numerical Descriptive Statistics on a daily basis by means of magnetic card
readers, bar code scanners, point of sale
• The most common numerical descriptive terminals, and touch screen monitors.
statistic is the mean (or average). • WalMart captures data on 20-30 million
• The mean demonstrates a measure of the transactions per day.
central tendency, or central location, of the • Visa processes 6,800 payment transactions
data for a variable. per second.
• Hudson's mean cost of parts, based on the 50 • Capturing, storing, and maintaining the
tune-ups studied, is $79 (found by summing up data, referred to as data warehousing, is a
the 50 cost values and then dividing by 50). significant undertaking.

Statistical Inference Data Mining

• Population - The set of all elements of interest in • Analysis of the data In the warehouse might
a particular study. aid in decisions that will lead to new
• Sample - A subset of the population. strategles and higher profits for the
• Statistical inference - The process of using data organization.
obtained from a sample to make estimates and • Using a combination of procedures from
rest hypotheses about the characteristics of a statistics, mathematics, and computer
population. science, analysts "mine the data" to convert
• Census - Collecting data for the entire it into useful information.
population. • The most effective data mining systems use
• Sample sur vey - Collecting data for a sample. automated procedures to discover
relationships in the data and predict future
outcomes, prompted by only general, even
vague, queries by the user.