Statistics for business decision making and analysis robert stine and foster chapter 01

1.1 What is Statistics?Some Basic Ideas  Statistics as a discipline  Questions we can answer with statistics  Variation, patterns and models... 1.1 What is Statistics?Statistics as a

Chapter 1

1.1 What is Statistics?

Some Basic Ideas

 Statistics as a discipline

 Questions we can answer with statistics

 Variation, patterns and models

1.1 What is Statistics?

Statistics as a Discipline

 Science and art of extracting answers from data

 A statistic is a property of data

 Statistics can be numbers (like an average) or

graphs that display information (like a timeplot)

1.1 What is Statistics?

Typical Questions Statistics Help to Answer

 Which customers are interested in an MP3

player?

 How much are customers willing to pay for an

MP3 player?

 Why does a shopper choose a particular box of

cereal?

1.1 What is Statistics?

Variation, Patterns and Models

 Variation refers to differences among items /

individuals or fluctuations over time

 Patterns are systematic or predictable features in data

 Models break down variation into a pattern and

unexplained variation

1.1 What is Statistics?

Statistical Models

 Key to extracting answers from data

 Good models simplify reality

 Finding real patterns helps us to understand data, plan for the future, and make better decisions

1.2 Previews

Predicting Employment

1.2 Previews

Predicting Employment

 Back in September 2005, how could you forecast employment in October?

 Use monthly employment data for U.S beginning January 2003

 Prepare a timeplot (a chart of values ordered in

time) to display monthly employment data

1.2 Previews

Timeplot of Employment Data

1.2 Previews

Linear Pattern and Region of Uncertainty

1.2 Previews

Conclusions

 Extrapolating the pattern (line) into the future

provides a forecast for October

 Actual employment for October is less than

expected and below the anticipated range

 This surprising result breaks from the pattern and deserves attention (the Katrina effect?)

1.2 Previews

Pricing a Car

1.2 Previews

Pricing a Car

 Is $22,000 too much to spend on a 2002 BMW

325i with 53,000 miles?

 Get prices for a sample of certified used BMW’s from the Philadelphia area in November 2005

 Prepare a scatterplot (a graph showing pairs of

values) of price versus mileage for this sample

1.2 Previews

Scatterplot of Price Versus Mileage

1.2 Previews

Line and Region of Uncertainty

1.2 Previews

Conclusions

 The range of uncertainty is so wide because

factors other than mileage affect price

 This model uses all of the data to answer the

question about one used car (borrowing strength)

 $22,000 isn’t too much to pay (it is less than

predicted by our line and within the range)

1.3 How to Use This Book

The Four M’s

 Motivation: reason for answering a question or

solving a problem

 Method: plan for addressing the problem

 Mechanics: performing the calculations,

generating graphs and displays

 Message: explanation of conclusions