Transcript Document

QM 2113 -- Fall 2003
Statistics for Decision Making
Descriptive Statistics
Instructor: John Seydel, Ph.D.
Student Objectives
Resolve questions, trouble, etc. regarding homework
exercises
Strengthen ability to do basic descriptive statistics


Calculations
Visual summaries
Recognize data types from primary source data
instruments
Know where to locate software for checkout


Office XP (in particular, Excel 2002)
WinXP
Calculate basic interval estimate for the population
mean
Perform basic informal bivariate analyses
First, Some Administrative Stuff
Questions about course?




Materials (e.g., syllabus, homework, . . . )
Policies (e.g., exams, grading, . . . )
Expectations
Other . . . ?
Collect homework, etc.



Prerequisite sheets
Charts for KIVZ data
Exercise 3-11
MIA: Campos, Gilbert, Harvey, Rose, Spragins
First exam is in 2 weeks (descriptive statistics)
Software Needed for Class
Office XP



Excel
PowerPoint
Word
If OfficeXP not feasible, get PowerPoint Viewer


Download from Microsoft site
Link on Handouts page of course website (untested)
Windows 98 or newer (ME, 2000, XP)
Where to get it:



Library (check out from circulation desk)
Download (only if you have broadband)
If problems, let me know
A Review
Quantitative data

Mean
 Just a simple average
 Add the values and divide by number of observations

Standard deviation
 Average difference among the values
 Process:





Subtract the average from each value
Square each result
“Average” the squared results
Take the square root of that result
Use a histogram to chart frequencies, relative frequencies
Qualitative data


Frequencies and relative frequencies
Use bar chart for charting
Let’s Look at the Homework
Chapter 1 questions


At this point, all you can do is quote the
text
However, these are good final exam
questions
Chapter 3:


Let’s look at #15
Any others?
How about other aspects of the
assignment?
Very Important: Data Type
Type of analysis depends upon data:

Quantitative; you’ll also see these terms
 Ratio
 Interval
 Ordinal

Qualitative; you’ll also see these terms
 Ordinal
 Nominal
General classifications of data



Information content
Time frame
Source
Consider a survey . . . (see handout)
A Quick Notation Quiz
Mean


Sample
Population
Standard deviation


Sample
Population
Variance
Range
Number of


Observations
Elements in the population
Miscellaneous Statistics
Less important but need to be familiar with:

Location

Variation
 Median
 Mode
 Quantiles (percentiles, quartiles)
 Range
 Min and Max
 CV

Both (?)
 Z-score
 Empirical Rule

Other: skew
Let’s take a look at these now . . .
Now, a Quick Peak at Inference
Two types


Estimation
Hypothesis testing
We’ll look closely at these later, but here’s
something to get things started
Estimating the population mean


Recall the Empirical Rule
Also, note that
 The sample mean is an unbiased estimator of the
population mean
 The standard deviation is a reasonable estimator of the
population standard deviation


That said, we can estimate m using x-bar . . .
This is called a confidence interval estimate
Bivariate Analyses: An
Introduction for Quantitative Data
Note: what we’ve been doing has dealt with a
single variable at a time


Summarizing its values
Describing its variation
Often, we want to explain that variation; we
seek to know why not all the observations
have the same value
That is, we seek to understand relationships
between two (or more) variables
Hence, bivariate (or multivariate) analysis is
called for
An Example: Great Northern Data
Look at Income (use Excel)


Average
Standard deviation
Why would it not be the same for all
adjustors?
Let’s examine how it relates to one of the
other variables; which?
Procedures:


Scatterplot
Estimate the relationship
Summary
Resolve questions, trouble, etc. regarding homework
exercises
Strengthen ability to do basic descriptive statistics


Calculations
Visual summaries
Recognize data types from primary source data
instruments
Know where to locate software for checkout


Office XP (in particular, Excel 2002)
WinXP
Calculate basic interval estimate for the population
mean
Perform basic informal bivariate analyses
Appendix
Sampling
Population
Sample
Statistic
Parameter
Schematic View
Statistics
Numeric Data
Informal
Summary Measures
Inferential Analyses
Categorical Data
Informal
Summary Measures
Inferential Analyses
Probability is what allows the linkage between descriptive and inferential
analyses
Work Expectations
Written work: type
Computational work:


Pencil, graph paper, straightedge
Computer printout
 Fit to page (when appropriate)
 Annotate with pencil as necessary
General guideline: be reasonable; e.g., if it
doesn’t lend itself to typing, do it manually or
with computer output
A Quick Overview
What is statistics all about?
It’s about dealing with variation


Summarizing information (description)
Making decisions based upon that summarization
Type of analysis depends on data type


Numeric
Categorical
Description

Formal
 Numeric data: average and standard deviaiton
 Categorical data: percentages

Informal: frequency tables and charts data