1
Inclass Practice 3
Topics covered  Hypothesis Testing and Correlation
(Bring the first 18 pages to class on 9/29 & 10/2)
NAME:
_______________________________________________________
BUAD 310
USC ID#
:__________________________________________________________
SIGNATURE
:___________________________________________________
ARIF ANSARI
P
a
r
t
1
MULTIPLE CHOICE
[2 points each]
1.
In statistical hypothesis testing, in order to determine the pvalue, which of the
following is not needed?
A.
The level of Significance
B.
Whether the test is lower tail, upper tail or two tail
C.
The Value of the test statistic
D.
The type of Distribution (Z or T)
2.
A significance test gives a pvalue of 0.04. From this we can
A.
Reject the null hypothesis at the 0.05 significance level.
B.
Reject the null hypothesis at the 0.01 significance level.
C.
Don’t reject the null hypothesis at the 0.05 significance level.
D.
Say that the probability that the null hypothesis is true is 0.04
3.
An office of a consulting firm prepares written and oral proposals to generate new
business.
The number of new proposals in any one week varies.
Summary
measures for the distribution are:
Oral Proposals
Written Proposals
Mean
2
2
Variance
1.83
1.65
Covariance
0.7
What is the correlation between the number of oral proposals in one week and the
number of written proposals in the week?
A.
1, because they are perfectly correlated.
B.
0, because the number of one kind of proposal doesn’t determine the
number of the other kind.
C.
0.70.
D.
Negative, and larger in magnitude than 0.70.
Page 1 of 49
This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document2
E.
Negative, and smaller in magnitude than 0.70.
4.
Generally speaking, if two variables are unrelated (as one increases, the other
shows no pattern), the correlation will be,
A.
A large positive number
B.
A large negative number
C.
A positive or negative number close to zero
D.
None of the above
5.
Generally speaking, if two variables are linearly related
and the magnitude of the
correlation is close to 1, we can say
a)
The two variable are correlated and there is a causal relationship
b)
The two variable not correlated but there is a causal relationship
c)
The two variable not correlated and there is no causal relationship
d)
The two variable correlated but there is a no causal relationship
e)
none of the above
Question 1 (6 points)
The following is a list of some of the techniques discussed in this course:
•
Confidence intervals
•
One sample T test
•
Histogram
•
Sampling distribution
•
Side by side boxplots
•
Scatterplot (Overlay plot)
For each of the situations described below, select the ONE technique from the preceding
list that you think is MOST applicable.
Write your answer in the space provided.
a.
You want to examine in detail the distribution of outstanding loans among
savings and loan institutions that went bankrupt in the S&L crisis of 1988.
You
have information collected at the firm level (one number per “bank”).
b.
You are a real estate broker in Manhattan Beach. As such, you are privy to certain
information about houses in MB.
You know that the average size (number of
square feet) of a house in Manhattan Beach is 2854 sq. ft.
You‘ve collected data
on 50 randomly selected houses in Santa Monica.
You want to know if houses in
Santa Monica, on average, are larger than 2854 square feet.
This is the end of the preview.
Sign up
to
access the rest of the document.
 Summer '07
 Lv
 Business, Statistics, Statistical hypothesis testing

Click to edit the document details