You are on page 1of 2

ENGM2032, Applied Probability and Statistics

Fall 2016

Assignment #9
Due: Tue Dec 6, 2016
1) A government agency wants to determine whether the mean radium-226 concentration in the
coolant discharge from a nuclear reactor is greater than the maximum allowable amount, 4
pCi/L. If they collect some discharge data, can they conclude that this is so?
a) Formulate the appropriate null and alternative hypotheses.
b) For this particular issue, what is the meaning of the type I error?
c) For this particular issue, what is the meaning of the type II error?
d) Which is the more serious error?
2) A company manufacturing certain laptop CPUs specifies that the mean power consumption of
the CPU does not exceed 40 W. You suspect that this claim is incorrect and randomly sample
twenty CPUs under normal operating conditions. The sample mean is observed to be 40.8
W. Suppose that it is known from past experience that the standard deviation of CPU power
consumption is 1.2 W. Based on your data, can you conclude with 95% confidence that the
specification provided by the company is false?
3) A company claims that the mean thickness of optical glass it produces is 5.0 mm. As part of
a daily quality control procedure, seven glass samples are selected and their thicknesses are
measured. Suppose that on a particular day, the measurements are:
4.71, 4.78, 5.02, 4.71, 4.90, 4.95, 5.01
If the glass thickness can be assumed to be at least approximately normally distributed, do the
data suggest that the mean thickness of the glass manufactured by the company on that day is
not 5 mm? Test at a 5% significance level.
4) There are concerns about the performance of a yogurt dispenser. In particular, there are reasons
to believe that the dispenser consistently underfills 50 ml containers. Nine filled containers are
randomly selected and the respective fill volumes are measured,
47.1, 44.8, 47.4, 51.9, 45.8, 47.3, 50.5, 46.3, 48.2.
Assuming that the fill volume is at least approximately normally distributed,
a) can you conclude with 95% confidence that the fill volume is below 50 ml?
b) what is the (approximate) p-value for this test? What does the magnitude of the p-value
tell you about the conclusion you have just made regarding the fill volume?
5) A company is currently purchasing its titanium alloy tie rods from supplier A. A new supplier
(supplier B) approaches the company and offers the same strength (at least according to supplier
Bs claim) tie rods at a lower price. The company is certainly interested in the offer. At the
same time, the company wants to make sure that the safety of their product is not compromised.
They will only switch to supplier B if they are sure that the mean yield strength of supplier Bs
rod is at least as large as that from supplier A. The company randomly selects ten tie rods from
each of the lots shipped by suppliers A and B and measures the yield strengths of the selected
rods. The observed sample mean and sample standard deviation are 651 MPa and 20 MPa for
supplier As rods, respectively, and the same parameters are 657 MPa and 23 MPa for supplier
Bs rods. Should the company consider switching to supplier B? Test at a 95% confidence
level.

ENGM2032, Applied Probability and Statistics

Assignment #9

Fall 2016: pg 2

6) Two crankshaft manufacturers, A and B, are being considered as suppliers for an engine. Both
manufacturers claim a mean fatigue life of 1.5 billion stress cycles. Because manufacturer A
uses a slightly superior machining process (at a slightly higher cost), the standard deviation
of the fatigue life for manufacturer A is 0.2 billion, while the standard deviation of fatigue
life for manufacture B is 0.25 billion. A random sample of 10 crankshafts is taken from each
manufacturer and fatigue lives of each measured, giving the following results (in billions);
xA = f1.8, 1.9, 2.0, 2.0, 1.7, 1.6, 1.1, 1.3, 1.7, 1.4g
xB = f1.7, 1.2, 1.1, 1.6, 1.4, 1.8, 1.9, 1.7, 1.6, 1.5g
a) Can it be concluded that the mean fatigue lives are different for the two manufacturers?
Test using = 0.05.
b) What is the p-value for this test?
7) Suppose that random sample of 400 Atlantic salmon found that 308 of them had fat contents
exceeding 10% while another sample of 360 Pacific salmon found that 240 had fat contents
exceeding 10%. Do these data indicate, with 80% confidence, that there is a significant
difference in the fractions of fish having fat contents in excess of 10% between the two oceans?
8) Consider the following two data sets.
set A:
x

0.1

0.14 0.33 0.62 0.75 1.22 1.45

1.52 1.85 1.81 2.03

2.6

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

set B:
x

1.2

1.4

2.80

1.64 0.83 2.42 1.53 2.84 1.65

1.6

1.8

2.0

2.2

2.4

2.8

3.0

3.2

3.4

1.91 2.23 2.86 2.67 0.44

For each of the data sets,


a) plot the data (y versus x). Are there any points that appear to be outliers? If there are,
circle and label them as such.
b) produce a regression of y against x. Superimpose the regression line on the plot produced
in step a). Do you think that the regression line captures the most important features of the
data set reasonably well?
c) using calculations at a 5% significance level, can you say that there is a significant linear
relationship between the x and y? That is, can you say with 95% confidence that y linearly
depends on x? Does this result agree with the conclusion you made in b)?
d) testing at a 5% significance level, can you say that the intercept (0 ) is not zero? How does
this conclusion agree with the plot in b)?

You might also like