BUSA320 Statistics for Decision-Making in Business Practice Exam 1
1. For each of the variables below, indicate its type (Categorical, Numerical Discrete or Continuous) and level of measurement (Ratio, Interval, Ordinal, Nominal): VARIABLE Color preferenc
Hypothesis Testing
Inference on one population proportion
Example
Conducting market research
Need at least 30% of produce buyers to be
willing to pay a premium for organic produce
A survey of regular customers is done
Out of 250 responses, 90 would be
Activity 1
Forming Teams and Sampling Review
BUAD 341
MOTIVATION
The predominant (but far from the only) mode of decision-making in statistics is thetest of signicance. The mechanics vary between situations, but the underlying logic is the same. we need
t
HYPOTHESIS TESTING USING AND (chapters 9,10 and 11)
Hypothesis testing (signicance testing) is used to decide whether sample information gives evidence of a
change or dierence in a population parameter. We set up a null hypothesis the statement of no chan
Outline
Hypothesis Test
known
unknown
Two Populations
Inference on the difference of two means, proportions
June 8, 2012
matched
Outline
Hypothesis Test
known
Outline
1
Hypothesis Test
2
Two Means, standard deviations known
3
Unknown standard deviation
Hypothesis Testing
General ideas
Hypotheses (plural of hypothesis)
Null: the fall-back case
Alternative: the active case. what you
want to test
Example: A manager typically assumes that
a 3 pound bag of coffee makes on average
at least 50 cups of coffee
Statistical Applications
ACTIVITY 5: Inference on Means - Two Populations
Why
Now that we have established the basic inferential methods condence intervals and tests of
signicance for means and proportions, we are ready to extend these to many other situa
Inference (hypothesis tests and estimation)
Dierences between means, proportions in two populations
The idea: We have two populations and a variable of interest. We may be interested in the dierence in mean values
(on this variable) in the two populations
Activity 9
Inference and prediction in multiple regression
BUAD 341
MOTIVATION
A regression equation describes the relation between a collection of predictors and the response.
This equation can help one draw conclusions about the the response variable in
Activity 8
Inference for Linear Regression
BUAD 341
MOTIVATION
For any set of points there is a line of best t, with a slope and an intercept (just as any set of
numbers has a mean and a standard deviation). for most really data, though, the point of coll
Table 1
Cumulative Probabilities for the Standard Normal Distribution
Entries in the table give
the area under the
curve to the left of the z
value. For example, for
z=-0.85, the cumulative
probability is 0.1977.
z
0.00
0.01
0.02
0.03
0.04
0.05
0.06
0.07
Quality Process
Control
Applying normal distributions to
production
What?
Testing your product as it is being
produced
Somewhat similar to lot acceptance
sampling, but you are not counting
defectives
You are measuring produced items to see
if they meet
Activity 6
Inference on two population variances
BUAD 341
MOTIVATION
It makes a lot of sense to compare the risk prole of two funds or the consistency of performance
of two processes (maybe manufacturing, audits, skiers, etc). We want to be able to use da
F-test
Inference on the ratio of two population variances
June 18, 2012
Inference on two population variances
What? Why? How?
I. Testing the variance, not the mean
Inference on two population variances
What? Why? How?
I. Testing the variance, not the mean
2 (chi-squared)
Inference on one population variance
June 18, 2012
Inference on one population variance
What? Why? How?
I. Testing the variance, not the mean
Inference on one population variance
What? Why? How?
I. Testing the variance, not the mean
II. Co
Lot
Acceptance
Sampling
Consumer and Producer Risk
Lot Acceptance Sampling
Producer: The company (or person) selling a
product.
You: The prospective purchaser.
Consumer: Your customer, the people (or
companies that buy products from you)
Standards for Lot
Poisson Random Variables
What? Discrete random variable. model the occurrence of rare events. is an event that occurs at most
10 times during a particular time interval.
Why? The Poisson random variable is a better model of the occurrence of rare events t
Activity 2
Discrete Random Variables
BUAD 341
MOTIVATION
The language of inferential statistics is based in probability and all our inferential formulas depend
on the distribution of a particular variable; the discrete variables give us our easiest approa
Activity 3
Continuous Random Variables
BUAD 341
MOTIVATION
Continuous random variables require dierent methods for calculation and description of their
distributions. The normal distributions provide the basis for most [but not all] of the inferential
sta
Regression Analysis
Simple linear regression
July 5, 2012
Simple Linear Regression
What? Why? How?
I. A way of nding a relationship between two numeric
variables
Simple Linear Regression
What? Why? How?
I. A way of nding a relationship between two numeric