Download Solutions to Exam 1 for Statistics for Engineers | STAT 4706 and more Exams Statistics in PDF only on Docsity! WK
> BOOM
Answ we Icey Form A
STAT 4706 - Statistics For Engineers
Exam #1
Fall 2008
Name:
Directions: You must do problem 1. Choose 3 out of the remaining 4 problems. If you do them all, only the
first three will be graded. Show all work.
Problem 1
Listed below are scenarios that need a specific confidence interval or hypothesis test. Select the letter from
the possible choices that corresponds to the appropriate interval or test and mark it to the left of the number
of the problem you're answering. Make sure to mark clearly on the exam paper. If the letter cannot be read
by me, then the problem will be marked wrong. (3 pts. each...you get 1 point free!)
A. One Sample z-test for G. One-Way Goodness of Fit Test.
B. One Sample t-test for H. Two-Way Test of Independence or Homogeneity
C. Two Independent Sample t-test | I. (1 — a) x 100% CI for w
D. Paired t-test J. (1—a) x 100% CI for pu — pe
E. Rank or Sign-Rank Test K. (1—a) x 100% CI for 6
F. Rank-Sum Test L. One or Two Sample Test for 7 or 7 — 72
We want to know if the average content of containers in a particular lubricant is 10 liters if we’re only
given measurements from a random sample of 10 containers and the distribution of contents is normal.
We have some data on the number of hours that two different types of scientific pocket calculators
operate before a recharge is required. We want to determine if calculator A operates longer than
calculator B on a full battery charge. Assume that the data does not come from a normal distribution.
“wo
oo
. A marketing expert for a pasta-making company believes that 40% of pasta lovers prefer lasagna. If
we counted the number of pasta lovers that prefer lasagna over other pastas from a random sample of
20 pasta lovers, we want to test if the expert’s claim is true.
. We have some data on grades in a statistics course for a particular semester. We want to test if the
distribution of grades is uniform.
on
. An independent consumer group tested radial tires from two major brands to determine whether there
were any differences in the expected tread life. Assume that each test car only has one brand of tires.
. Some data was recorded that represents the length of time to recovery for patients randomly treated
with one & two medications to clear up severe bladder infections. We want to estimate the difference
in mean recovery time for the two medications.
=
. The ocean swell produces spectacular eruptions of water through a hole in the cliff at Kiama, Australia,
known as the Blowhole. Historically, the time between eruptions has been 60 seconds with a standard
deviation of 10 seconds. The researcher wants to determine whether the true mean time between
eruptions has increased from a random sample of the times at which 10 successive eruptions occurred.
. In astudy conducted by the Department of Human Nutrition and Foods at VPI&SU, data was recorded
on the comparison of sorbic acid residuals in parts per million in ham immediately after dipping in
a sorbate solution and after 60 days of storage were recorded. We want to know if there’s sufficient
evidence to say that the length of storage influences sorbic acid residual concentrations.
oo
Problem 2
The following data represent the running times (in minutes) of films produced by 2 motion-picture companies:
Company 1 | Company 2
102 81
86 165
98 97
109 134
92 92
87
114
Test the hypothesis that the average running time of films produced by company 2 exceeds the average
running time of films produced by company 1 by 10 minutes, against the alternative that the difference is
not 10 minutes.
a) (5 pts.) State the appropriate hypotheses for this test. You may purchase this answer for 5 points to
make sure that the rest of the problem can be done correctly. Come and see me to purchase.
Noi -Mr = he
He ‘ay -A2 wat
) (20 pts.) First, construct and interpret a 90% confidence interval for 4; — jig, and then use that interval
to test the hypotheses in part a). Assume that s; = a 88 = 89 = 30.2. 2 <= IO
(30-2) es
are (S/n. + 5*Mn2)? (Ce Ye+
sous * (EA)? (eo)? = ed aed
in.) SZinde _sSae/n,): ie
G = a S ey
Qe at oe
2 = See Ae
_ 10 ees = a (~¥2 yee
(%-88)* _ ay
= @44-uo) +t a as |e
1,64S
(ie Oe ay 10, ORS
Sie ane cencameepeete eects
flrs produced hg Conperty C CxcedoYk avrag
7 tine Une Aoorce”) b [ 4
pees tice op fA doe, Foil delee Uo sh
Pied ica ee anna to Yur - 35.5023 § py, 0M roy,
*< | ><\
Problem 5
A food inspector examines 16 jars of a certain brand of jam to determine the percent of foreign impurities,
A test for normality was done to see whether a nonparametric test was appropriate or not. Consider the
following output from Minitab.
Probability Plot of PctImpurities
Normal
wean 287
Site 08747
N 6
Ks 027
Pevalue 0.086
a) (5 pts.) Given the above graph and a 5% significance level, should we be using a nonparametric test?
Plain vere Cok, so we rye Ho And cénarccy fa
ea dota Aorran 4% Care bem a mevmal. dist.
> oe a Nerporomelice fet
b) (20 pts.) Now consider some Minitab output from a nonparametric test. Which parametric test could
we have done instead of this? And test if the median percent of impurities in this brand of jam is not
2.5% with the same level of significance.
Wilcoxon Signed Rank Test: PctImpurities
GQ Ere.r sanphe £~ fet
Test of median = 2.500 versus median not = 2.500 Unghted
N for Wilcoxon Estimated
Bah
N Test Statistic B~ fedien HO f= 25°
PctImpurities 16 16 35.5 0.098 2.100 Hen BF £ 2 Sv
CH= OS.
P-VehR = 095 >&
> we fal do ayot Fs rena Conde phat te true
jrradian prereset of Vrpuriles Co LS fy RS
Cn PAIRED COMPARISON EXPERIMENT (6 ONE OF THE MOST EFFECTIVE WAYS TO |
REDUCE MATURAL VARIABILITY WHILE COMPARING TREATMENTS, FOR EXAMPLE, IN
COMPARING HAND CREWAS, THE TWO BRANDS ARE RANDOMLY ASSIGNED TO
GACH SUBJECT’S RIGHT OR LEFT WANDS THIS ELIMINATES VARIABILITY QUE TO
SKIN DIFFERENCES.
= (P40 2 ue wos Ox |