Scribd
Upload
360 views

Tutorial 08 Solutions

The document provides instructions and examples for using Excel to calculate probabilities and critical values associated with the Student's t distribution. It includes four examples involving calculating confidence intervals for means using data from samples. The examples involve determining confidence intervals for the mean width of steel troughs produced by a manufacturing company and for the mean price of petrol in a city.

Uploaded by

cynthia
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
360 views

Tutorial 08 Solutions

The document provides instructions and examples for using Excel to calculate probabilities and critical values associated with the Student's t distribution. It includes four examples involving calculating confidence intervals for means using data from samples. The examples involve determining confidence intervals for the mean width of steel troughs produced by a manufacturing company and for the mean price of petrol in a city.

Uploaded by

cynthia
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 8

Tutorial 08: Interval Estimation

Part A: Use Excel for Part A


For all your answers, please remember to do the following:
1. Draw curves
2. State the distributionl
3. Define the variable

A8.1
Probabilities and critical values associated with the Student t distribution can
be found using the Excel functions T.DIST, T.DIST.2T, T.DIST.RT, and T.INV,
T.INV.2T, as discussed in Section E4.5 of the Excel notes (refer to Software
Information folder on Moodle). Work through Section E4.5 and then answer
the following questions, writing your answer in the space provided below
each question part.
If t is drawn from the Student t distribution with 19 degrees of freedom,
(a) find the probability P(t> 2) using T.DIST.RT
T.DIST.RT(2,19)=0.030001
(b) find the probability P( t<-2) using T.DIST
T.DIST(- 2, 19, TRUE) = 0.030001

(c) find the probability P( t>2 or t<-2) using T.DIST.2T


T.DIST.2T( 2, 19) = 0.060002
(d) Find the critical value of tcrit such that P(t<tcrit)= 0.02 = using T.INV
T.INV(0.02,19) = -2.2047

(e) Find the value tcrit t of t such that P(t>tcrit)= 0.02 = using your answer
to (d).
Based on (d), we conclude that tcrit=2.2047
1

(f) Find the critical value tcrit of t such that P(t>tcrit or t<-tcrit)= 0.03 using
T.INV.2T
T.INV.2T(0.03,19) = 2.3456

A8.2
A manufacturing company produces steel housings for electrical equipment. The
main component of the housing is a steel trough made out of a 2mm steel coil. The
width of the troughs is critical because of weatherproofing in outdoor situations.
In T8.xls the widths of a sample of 49 troughs are listed. Use Excel to find the mean of
the sample, and assuming that the population standard deviation is 0.46mm,
calculate a 95% confidence interval for the mean width of the troughs. Show your
answers to three decimal places. Show ALL your working.

Let X=width of trough (mm)


Assume

X ~ N , 0.46 2

95%

0.46

n = 49, X 204.209

Z:

1-=0.95, so =0.05,

-1.96

1.96

Z 0.05 Z 0.025 1.96


2

The 95% confidence interval for :

X Z
X Z

n
n
2
2

204.209 1.96X

0.46
0.46
204.209 1.96X
49
49

204.209 1.96X0.0657 204.209 1.96X0.0657


204.209 0.1288 204.209 0.1288
204.0802 204.3378

Interpretation:
We are 95% confident that the true mean width of the steel troughs lies between
204.08mm and 204.34mm
2

A8.3
A manufacturing company produces steel housings for electrical equipment.
The main component of the housing is a steel trough made out of a 2-mm
steel coil. The width of the troughs is critical because of 2 weatherproofing in
outdoor situations. In T8.xls the widths of a sample of 49 troughs are listed. Use
Excel to find the appropriate critical values, sample mean and standard
deviation.
Calculate a 95% confidence interval for the mean width of the troughs. State
the formula, show all working and remember to always interpret your interval
in the context of this question.

A8.4 (Based on Berenson 8.13 page 253)


Consider the following sets of data drawn from a normally distributed
population.
Set A: 1,1,1,1,8,8,8,8
Set B: 1,2,3,4,5,6,7,8
Each set of data is used to calculate a 95% confidence interval for the
population mean. Without doing any calculations, state, with explanation
which confidence interval will be wider. What can you say then, about the
precision of estimate?
The confidence interval will be wider if the variance is higher, that is, if the
average squared distance from the mean is higher. Both data sets have a
mean of 4.5, and Set A has more data points that are far from the mean.
Therefore Set A will result in a wider confidence interval.
A wider interval results in a decrease in the precision of estimate. Therefore,
the confidence interval based on Set A will be LESS precise than that based
on Set B.

Part B: Use tables for all of the following question


B8.5
(a) Comment on the difference between the answers to A8.2 and A8.3
(above).
The answers differ only slightly. To two decimal places, they are identical. This
can be explained by
(1) The fact that the sample size is reasonably large so that the t distribution is
not very different from the standard normal distribution (the critical value
is 2.01 instead of 1.96).
(2) The sample standard deviation happens to be very close to the
population standard deviation.
(b) If you were attempting question A8.3 without the use of Excel, what
degrees of freedom would you use when consulting the table? How

inaccurate do you think your answer might be?


The correct degrees of freedom are 48. 45 and 50 degrees of freedom are
provided in the table, and the corresponding critical values are 2.014 and
2.009 respectively. To 3 decimal places, the correct critical value for 48
degrees of freedom is 2.011, a difference of 0.003 from the 45 degrees of
freedom answer and 0.002 from the 50 degrees of freedom answer. For this
particular problem, either of these will result in a change to the confidence
limits that makes a difference at the 4th decimal place.

B8.6
A random sample of 20 petrol stations in the city of Casey on a Tuesday found that
the mean price per litre over the 20 stations was $1.52. Assume that the price of
petrol is normally distributed.

(a) Assuming the population standard deviation was 3 cents, find a 95%
confidence interval for the mean price of unleaded petrol in Casey on that
day.
Let X=price per litre of petrol ($)
Assume

X~N

0.032

X 1.52
0.03
n 20
1 0.95 0.05

z z 0.025 1.96

95%


X Z
X Z 2

2
n
n

1.52 1.96X

Z:

-1.96

1.96

0.03
0.03
1.52 1.96X
20
20

1.52 0.0132 1.52 0.0132


1.5068 1.5331
5

Assuming the prices are normally distributed, we can state with 95%
confidence that the mean price in Casey on a Tuesday is between $1.51 and
$1.53.
(b) To obtain a 90% confidence interval, just need to change the critical value to

X 1.52
0.03
n 20
1 0.90 0.10

z z 0.05 1.645
2

95%
Z:


X Z
X Z 2

2
n
n

1.52 1.645X

-1.645

1.645

0.03
0.03
1.52 1.645X
20
20

1.52 0.011 1.52 0.011


1.5090 1.5310
Assuming the prices are normally distributed, we can state with 90%
confidence that the mean price in Casey on a Tuesday is between $1.51 and
$1.53.

(c) If n 80 , in other words the sample is 4 times as big, but the mean and other
details are the same, then:
the width of the interval will be more narrow (see note below)
which results in an increase in precision of the estimate.

1.52 0.0066 1.52 0.0066


1.5134 1.5266
Note: the half width depends on

1
and
n

1
1 1
, so the half width is

80 2 20

halved. So the half width for a 95% confidence interval is now half of 0.01315, or
0.0066.

B8.7
With the recent new packaging legislation, there is a lot of interest at the
moment in the proportion of Australians who smoke. Suppose a survey is
conducted on the smoking habits of 5,000 randomly chosen Australians aged
15 years or older. It is found that 784 of the participants in the survey currently
smoke.
Obtain a 95% confidence interval for the proportion of Australians aged over
15 who currently smoke. State the formula, show all working and remember to
always interpret your interval in the context of this question.

Let X = No of Australians, 15 yrs or older, who smoke.

X=784, n=5,000

np=5000X0.1568=784 and n(1-p)=5000(1-0.1568)=4216


Since both are greater than 5, the normal approximation is valid.


With 95% confidence, we can state that the proportion of Australians over 15 years
of age who currently smoke is between 0.147 and 0.167 Or, 14.7% - 16.7%

You might also like