Vous êtes sur la page 1sur 5

Assignment 5

Spring Semester 2019


ES 202 - Engineering Statistics

1. A random sample of size n is taken from the Poisson pdf,


e−λ λx
; x ≥ 0.
fX (x; λ) =
x!
(a) Find the MME (method of moments estimate) for λ.
(b) Find the MLE (maximum likelihood estimate) for λ.
1P 1P
Answer: (a) λ̂m = xi = x̄ (b) λ̂l = xi = x̄
n n
2. Use the method of moments to estimate θ in the pdf
fY (y; θ) = (θ2 + θ)y θ−1 (1 − y); 0 ≤ y ≤ 1.
Assume that a random sample of size n has been collected.
P
2 yi 2ȳ
Answer: θ̂m = P =
n − yi 1 − ȳ
3. An engineer is creating a project scheduling program and recognizes that the tasks
making up the project are not always completed on time. However, the completion
proportion tends to be fairly high. To reflect this condition, he uses the pdf
fY (y; θ) = θy θ−1 ; 0 ≤ y ≤ 1, and θ > 0,
where y is the proportion of the task completed. Suppose that in his previous
project, the proportions of tasks completed were 0.77, 0.82, 0.92, 0.94, and 0.98.
(a) Find the MME (method of moments estimate) for θ.
(b) Find the MLE (maximum likelihood estimate) for θ.
P
y ȳ −n
Answer: (a) θ̂m = Pi = = 7.7719 (b) θ̂l = P = 7.9965
n − yi 1 − ȳ ln yi
4. Suppose an isolated weather-reporting station has an electronic device whose time
to failure is given by the exponential model
1
fY (y; θ) = e−y/θ ; 0 ≤ y < ∞; 0 < θ < ∞.
θ
The station also has a spare device, so the time until this instrument is not available
is the sum of these two exponential pdfs, which is
1 −y/θ
fY (y; θ) = ye ; 0 ≤ y < ∞; 0 < θ < ∞.
θ2
Five data points have been collected: 9.2, 5.6, 18.4, 12.1, and 10.7. Find the maxi-
mum likelihood estimate for θ.
1 P 1
Answer: θ̂l = yi = ȳ = 5.6
2n 2
1
5. Use the method of maximum likelihood to estimate θ in the pdf
θ √
fY (y; θ) = √ e−θ y ; y ≥ 0.
2 y

Evaluate θ̂l for the following random sample of size 4: Y1 = 6.2, Y2 = 7.0, Y3 = 2.5,
and Y4 = 4.2.

n n
r
Answer: θ̂l = pP = = 0.8967
yi ȳ

6. Suppose that Y1 = 8.3, Y2 = 4.9, Y3 = 2.6, and Y4 = 6.5 is a random sample of size
4 from the two-parameter uniform pdf,
1
fY (y; θ1 , θ2 ) = ; θ1 − θ2 ≤ y ≤ θ1 + θ2 .
2θ2
Use the method of moments to estimate θ1 and θ2 .
r
(yi − ȳ)2
P P
yi 3×
Answer: θ̂1m = = ȳ = 5.575 θ̂2m = = 3.6319
n n
7. Weibull distribution measures time to failure of devices where the probability of
failure increases as time does. A Weibull random variable Y has pdf
β
fY (y; α, β) = αβy β−1 e−αy ; y ≥ 0, α > 0, β > 0.

(a) Find the maximum likelihood estimator for α assuming that β is known.
(b) Suppose α and β are both unknown. Write down the equations (but do not
solve them) that would be solved simultaneously to find the maximum likeli-
hood estimators of α and β
n
Answer: (a) α̂l = P β (b) Equations are:
yi
n X β n X X β
− yi = 0 and + ln yi − α yi ln β = 0
α β

8. The shopping times of n = 64 randomly selected customers at a local supermarket


were recorded. The average and variance of the 64 shopping times were 33 minutes
and 256 minutes2 , respectively. Estimate µ, the true average shopping time per
customer, with a confidence coefficient of 1 − α = .90.

Answer: 33 ± 3.284 = (29.716, 36.284)

9. In a Gallup Poll of n = 800 randomly chosen adults, 45% indicated that movies
were getting better whereas 43% indicated that movies were getting worse. Find a
98% confidence interval for p, the overall proportion of adults who say that movies
are getting better.

Answer: 0.45 ± 0.041 = (0.409, 0.491)

2
10. One suggested method for solving the electric-power shortage in a region involves
constructing floating nuclear power plants a few miles offshore in the ocean. Con-
cern about the possibility of a ship collision with the floating (but anchored) plant
has raised the need for an estimate of the density of ship traffic in the area. The
number of ships passing within 10 miles of the proposed power-plant location per
day, recorded for n = 60 days during July and August, possessed a sample mean
and variance of ȳ = 7.2 and s2 = 8.8. Find a 95% confidence interval for the mean
number of ships passing within 10 miles of the proposed power-plant location dur-
ing a 1-day time period.

Answer: 7.2 ± .751 = (6.449, 7.951)

11. The EPA has set a maximum noise level for heavy trucks at 83 decibels (dB). A
random sample of six heavy trucks produced the following noise levels (in decibels):

85.4 86.8 86.1 85.3 84.8 86.0

Use these data to construct a 90% confidence interval for σ 2 , the variance of the
truck noise-emission readings.

Answer: (0.227, 2.187)

12. A precision instrument is guaranteed to read accurately within 2 units. A sample


of four instrument readings on the same object yielded the measurements 353, 351,
351, and 355. Find a 90% confidence interval for the population variance. Does the
guarantee seem reasonable?

Answer: (1.408, 31.429); No

13. A chemist who has prepared a product designed to kill 60% of a particular type of
insect wants to evaluate the kill rate of her preparation. What sample size should
she use if she wishes to be 95% confident that her experimental results fall within
0.02 of the true fraction of insects killed?

Answer: 2305

14. Past experience shows that the standard deviation of the yearly income of textile
workers in a certain state is $400. How many textile workers would you need to
sample if you wished to estimate the population mean to within $50, with proba-
bility 0.95?

Answer: 246

15. A chemical process has produced, on the average, 800 tons of chemical per day.
The daily yields for the past week are 785, 805, 790, 793, and 802 tons.

(a) Estimate the mean daily yield, with confidence coefficient 0.90, from the data.
(b) Find a 90% confidence interval for σ 2 , the variance of the daily yields.

Answer: (a) (788.867, 801.133) (b) (29.294, 391.549)

3
16. A mathematics test is given to a class of 50 students randomly selected from high
school 1 and also to a class of 45 students randomly selected from high school 2. For
the class at high school 1, the sample mean is 75 points, and the sample standard
deviation is 10 points. For the class at high school 2, the sample mean is 72 points,
and the sample standard deviation is 8 points. Construct a 95% confidence interval
for the difference in the mean scores.

Answer: 3 ± 3.626 = (−0.626, 6.626)

17. Two brands of refrigerators, denoted A and B, are each guaranteed for 1 year. In
a random sample of 50 refrigerators of brand A, 12 were observed to fail before
the guarantee period ended. An independent random sample of 60 brand B refrig-
erators also revealed 12 failures during the guarantee period. Estimate the true
difference (p1 − p2 ) between proportions of failures during the guarantee period,
with confidence coefficient approximately 0.98.

Answer: 0.04 ± 0.1851 = (−0.145, 0.225)

18. Refer to previous problem 17.

(a) At the approximate 98% confidence level, what is the largest “believable value”
for the difference in the proportions of failures for refrigerators of brands A
and B?
(b) At the approximate 98% confidence level, what is the smallest “believable
value” for the difference in the proportions of failures for refrigerators of brands
A and B?
(c) If p1 − p2 actually equals 0.2251, which brand has the larger proportion of
failures during the warranty period? How much larger?
(d) If p1 − p2 actually equals -0.1451, which brand has the larger proportion of
failures during the warranty period larger? How much larger?
(e) As observed in Problem 17, zero is a believable value of the difference. Would
you conclude that there is evidence of a difference in the proportions of failures
(within the warranty period) for the two brands of refrigerators?

Answer: (a) 0.225 (b) zero (c) Brand A has 22.5% more failures
than brand B (d) Brand A has 14.5% less failures than brand B (e) No
conclusion can be drawn from this observation. Difference in proportion of failures
for the two brands may or may not exist.

19. An experimenter wanted to check the variability of measurements obtained by using


equipment designed to measure the volume of an audio source. Three independent
measurements recorded by this equipment for the same sound were 4.1, 5.2, and
10.2. Estimate σ 2 with confidence coefficient 0.90.

Answer: (3.529, 211.4)

20. The ages of a random sample of five university professors are 39, 54, 61, 72, and 59.
Using this information, find a 99% confidence interval for the population standard
deviation of the ages of all professors at the university, assuming that the ages of

4
university professors are normally distributed.

Answer: (6.249, 52.463)

21. A random sample of size 25 was taken from a normal population with σ 2 = 6. A
confidence interval for the mean was given as (5.37, 7.37). What is the confidence
coefficient associated with this interval?

Answer: 1 − 2 × (0.0207) = 0.9586

22. To estimate the proportion of unemployed workers in Panama, an economist se-


lected at random 400 persons from the working class. Of these, 25 were unem-
ployed.

(a) Estimate the true proportion of unemployed workers and place bounds on the
error of estimation.
(b) How many persons must be sampled to reduce the bound on the error of
estimation to 0.02?

Answer: (a) 0.0625 ± 0.0237 = (0.0388, 0.0862) (b) 563

23. How many voters must be included in a sample collected to estimate the fraction of
the popular vote favorable to a presidential candidate in a national election if the
estimate must be correct to within 0.005? Assume that the true fraction lies some-
where in the neighborhood of 0.5. Use a confidence coefficient of approximately 0.95.

Answer: n = 38, 416

24. The annual main stem growth, measured for a sample of 17 4-year-old red pine
trees, produced a mean of 11.3 inches and a standard deviation of 3.4 inches. Find
a 90% confidence interval for the mean annual main stem growth of a population
of 4-year-old red pine trees subjected to similar environmental conditions. Assume
that the growth amounts are normally distributed.

Answer: 11.3 ± 1.44 = (9.86, 12.74)

25. A pharmaceutical manufacturer purchases raw material from two different suppli-
ers. The mean level of impurities is approximately the same for both suppliers, but
the manufacturer is concerned about the variability in the amount of impurities
from shipment to shipment. If the level of impurities tends to vary excessively for
one source of supply, this could affect the quality of the final product. To compare
the variation in percentage impurities for the two suppliers, the manufacturer se-
lects ten shipments from each supplier and measures the percentage of impurities
in each shipment. The sample variances were s21 = 0.273 and s22 = 0.094, respec-
tively. Form a 95% confidence interval for the ratio of the true population variances.

Answer:

Vous aimerez peut-être aussi