Statistics 3N03 - Assignment #3

2000-11-20

Due: 2000-12-01 17:00


You may use any calculators or software you like, but you will need to be able to do problems like these under exam conditions, so if you have time you should do them on your calculator and then check your results in MINITAB, SPSS or R.

Text references are to Montgomery & Runger, Applied Statistics and Probability for Engineers, 2nd edition.

Part A

1. Figure 9.4 on page 411 is inaccurate and misleading. Draw it correctly.

2. How many independent observations are needed to ensure that the upper limit of a 95% confidence interval for the variance of a normal population is no more than 4 times its lower limit?

3. Ten patients were given a drug intended to reduce diastolic blood pressure, which was recorded before and after treatment. Compute a P-value to help determine if treatment was effective. State your assumptions and conclusions.

Patient:

1

2

3

4

5

6

7

8

9

10

Before:

120

125

131

115

100

105

117

93

112

116

After:

116

123

140

108

91

103

102

109

111

113

4. Ten patients were given a drug intended to reduce diastolic blood pressure, but the only information you are given is that diastolic blood pressure dropped in 8 out of the 10 patients. Compute a P-value to help determine if treatment was effective. State your assumptions and conclusions. Which analysis is the more valid, this or the t-test in Question 3? [Hint: What is the distribution of the number of patients with reduced diastolic blood pressure, under the null hypothesis? Follow the definition of P-value exactly.]

Part B

Problems from the text.

8-39

9-18 (d), 9-19*, 9-36, 9-52

10-12, 10-27

* In 9-19, do stem-and-leaf plots and a comparative box plot instead of probability plots, and do an F-test to compare the variances.


Statistics 3N03