Final Exam with Solution - Statistical Methods I | STAT 515, Exams of Data Analysis & Statistical Methods

Material Type: Exam; Class: STATISTICAL METHODS I; Subject: Statistics; University: University of South Carolina - Columbia; Term: Fall 2001;

Typology: Exams

Pre 2010

Uploaded on 09/02/2009

koofers-user-ixs
koofers-user-ixs šŸ‡ŗšŸ‡ø

10 documents

1 / 2

Toggle sidebar

This page cannot be seen from the preview

Don't miss anything!

bg1
STAT 515 – Fall 2001 - Final Exam - Solutions
Part I: Answer three of the following four questions. If you answer more than three I will grade only the first three.
Five points each.
1) Define what is meant by the p-value (or observed significance level) of a test. The p-value is the probability of
observing a statistic as extreme as the one observed, or more extreme, if the null hypothesis is true. -or- The p-
value is the smallest α
αα
α-level at which the null hypothesis would be rejected.
2) The histogram at the right is skewed left We would expect that its mean would be smaller than its median.
3) A fair coin (probability of a head on one flip = 0.5) is flipped 10 times. What is the probability of observing exactly
5 heads? 2461.0)000976562.0(252
10
5.0
!5!5
!10
5
)5.01(
5
5.0
5
10 ā‰ˆ==āˆ’






4) A fair coin (probability of a head on one flip = 0.5) is flipped 1,000 times. Using the normal approximation to the
binomial, approximately what is the probability of observing at least 512 heads?
73.0
811.15
5.11
)5.01)(5.0(1000
)5.0(1000)5.0512(
)1(
)( ā‰ˆ=
āˆ’
āˆ’āˆ’
=
āˆ’
āˆ’
=pnp
npX
Z
We need to subtract the 0.5 for the continuity correction. The answer we want is P(Z>0.73) = 0.5 -
0.2673=0.2327
Part II: Answer every part of the next three problems. Read each problem carefully, and show your work for full
credit. Twenty points each.
1A) State the appropriate null and alternate hypothesis to test if a high score on the exam seems to indicate mild to
moderate traumatic brain injury. Identify any parameters you use in stating the hypotheses.
H0: µ
µµ
µ=40 vs. HA: µ
µµ
µ>40 where µ
µµ
µ is the average test score of all patients with mild to moderate traumatic brain
injuries
B) Test the hypothesis in part a. Report your conclusion at α=0.05 (e.g. Do we accept or reject H0? Is a high test
score associated with traumatic brain injury, or not?)
compare this to 1.717 and reject H0.
The injured do seem to score higher.
C) Construct a 95% confidence interval for the mean test score of patients who suffered mild to moderate traumatic
brain injury.
)41.57,45.39(98.843.48
23
76.20
074.243.48
2=±=±=± n
s
tx
α
D) What assumptions need to be satisfied in order to trust the results you obtained in parts B and C?
The patients need to be a random sample and their test scores need to follow a normal distribution.
95.1
32.4
43.8
23
76.20
4043.48
221 ==
āˆ’
=
āˆ’
=
=āˆ’=
n
s
x
ndf
t
µ
pf2

Partial preview of the text

Download Final Exam with Solution - Statistical Methods I | STAT 515 and more Exams Data Analysis & Statistical Methods in PDF only on Docsity!

STAT 515 – Fall 2001 - Final Exam - Solutions

Part I: Answer three of the following four questions. If you answer more than three I will grade only the first three. Five points each.

  1. Define what is meant by the p-value (or observed significance level) of a test. The p-value is the probability of observing a statistic as extreme as the one observed, or more extreme, if the null hypothesis is true. -or- The p- value is the smallest αααα -level at which the null hypothesis would be rejected.

  2. The histogram at the right is skewed left We would expect that its mean would be smaller than its median.

  3. A fair coin (probability of a head on one flip = 0.5) is flipped 10 times. What is the probability of observing exactly

5 heads? 252 ( 0. 000976562 ) 0. 2461

  1. A fair coin (probability of a head on one flip = 0.5) is flipped 1,000 times. Using the normal approximation to the binomial, approximately what is the probability of observing at least 512 heads?

np p

X np Z

We need to subtract the 0.5 for the continuity correction. The answer we want is P(Z>0.73) = 0.5 - 0.2673=0.

Part II: Answer every part of the next three problems. Read each problem carefully, and show your work for full credit. Twenty points each.

1A) State the appropriate null and alternate hypothesis to test if a high score on the exam seems to indicate mild to moderate traumatic brain injury. Identify any parameters you use in stating the hypotheses. H (^) 0: μμμμ =40 vs. H (^) A : μμμμ >40 where μμμμ is the average test score of all patients with mild to moderate traumatic brain injuries

B) Test the hypothesis in part a. Report your conclusion at α=0.05 (e.g. Do we accept or reject H0? Is a high test score associated with traumatic brain injury, or not?)

compare this to 1.717 and reject H 0. The injured do seem to score higher.

C) Construct a 95% confidence interval for the mean test score of patients who suffered mild to moderate traumatic brain injury.

n

s

x t α

D) What assumptions need to be satisfied in order to trust the results you obtained in parts B and C?

The patients need to be a random sample and their test scores need to follow a normal distribution.

n

s

x tdf n

2A) Note that four values have been deleted from the ANOVA table. What values should they have?

DF for total _____________ MS for Error __________ MS for model _____________ F Statistic __________

B) Assuming the assumptions for predicting this linear regression are met, perform the test of hypotheses for testing

whether batting average predicts wins. What is the p-value, and what is your conclusion α=0.05?

The p-value is 0.0014. We reject the null hypothesis and conclude that batting average does predict wins.

C) Assuming the assumptions for this simple linear regression are met, identify which of the following statements are

true or false.

F 58.8% of a team’s wins are explained by batting average.

T 58.8% of the variation in the number of wins a team has is explained by the batting average.

F The correlation coefficient for this regression is 0.588.

T The coefficient of determination for this regressions is 0.588.

D) Assuming the assumptions for this simple linear regression are met, give the estimated range that the model says

95% of the teams batting 0.300 should have win totals between. From the prediction interval, 85.9738 to 136.8921.

E) Why can’t we trust the interval in D? A 0.300 batting average is out of the range of our data (we would be

extrapolating.)

  1. Male Viewer Female Viewer Identified Product 95 41 136

Could Not Identify Product 55 109 164 150 150 300

A) Would this data be analyzed by using a test of independence, a test of homogeneity , or a goodness of fit test?

B) Write out the tables of expected values for conducting this test.

Male Viewer Female Viewer Identified Product (136)(150)/300=68 (136)(150)/300=

Could Not Identify Product (150)(164)/300=82 (150)(164)/300=

C) Give the formula for X 2 for this problem (plugging the values in, but not needing to simplify).

2 (^9568 ) āˆ’

X =

D) What is the rejection region (critical region) for conducting this test at α=0.05? df=1, so Reject if X 2 ≄3.84146.

E) Why is, or why isn't, the sample size of this experiment large enough for performing this hypothesis test?

It is large enough because all of the expected values are at least 5.

Summary of Fit Mean of Response 81. Root MSE 8.

R-Square 0. Adj R-Sq 0.

Analysis of Variance Source Model Error C Total

DF 1 12 13

Sum of Squares

Mean Square 1322. 77.

F Stat 17.

Pr > F