To select: A random sample of 8 storms by using random numbers.
To find: The average maximum wind speed.
To compare: The results with the population

Answer to Problem 14.1.1RE
The random sample of 8 storms by using random numbers is:
Serial No. | Name |
01 | Arlene |
06 | Franklin |
12 | Lee |
13 | Maria |
14 | Nate |
18 | Stan |
19 | Unnamed |
27 | Epsilon |
The average maximum wind speed of the sample is 75.
By comparing the sample mean of 75 with the population mean of 87.5, it is clear that there is some difference between the sample mean and the population mean.
Explanation of Solution
Given info:
In the 2005 Atlantic hurricane season the maximum wind speed and classifications for different storms are given in the data.
Calculation:
Answers will vary. One of the possible answers is given below:
First arrange the data according to the serial number,
Serial No. | Name | Max. wind | Classification |
1 | Arlene | 70 | Storm |
2 | Bret | 40 | S |
3 | Cindy | 75 | Hurricane |
4 | Dennis | 150 | H |
5 | Emily | 160 | H |
6 | Franklin | 70 | S |
7 | Gert | 45 | S |
8 | Harvey | 65 | S |
9 | Irene | 105 | H |
10 | Jose | 50 | S |
11 | Katrina | 175 | H |
12 | Lee | 40 | S |
13 | Maria | 115 | H |
14 | Nate | 90 | H |
15 | Ophelia | 85 | H |
16 | Philippe | 70 | H |
17 | Rita | 175 | H |
18 | Stan | 80 | H |
19 | Unnamed | 50 | S |
20 | Tammy | 50 | S |
21 | Vince | 75 | H |
22 | Wilma | 175 | H |
23 | Alpha | 50 | S |
24 | Beta | 115 | H |
25 | Gamma | 55 | S |
26 | Delta | 70 | S |
27 | Epsilon | 85 | H |
28 | Zeta | 65 | S |
From the data select a random sample of 8 storms by using random numbers.
Procedure for selecting eight random samples:
- From the figure 14.1(Table of random numbers), select a starting point.
- Here, the selected starting point is ‘18’ which is present in third row and first column. There are 28 names of storms. Hence, select the random numbers in the
range of ‘1’ to ‘28. Avoid repetition of numbers and 00. - The random numbers starting from ‘18’ are:
18, 19, 14, 29, 01, 55, 84, 62, 66, 48, 94, 100, 46, 77, 81, 40, 41, 52, 13, 82, 57, 12, 27, 75, 95, 62, 57, 13, 31, 06.
- Select 18, 19, 14 which are in the range (01 to 28).
- Next number 29 is not selected because it is not in the range (1 to 28).
- And select 01.
- Repeat the same until eight random numbers are obtained.
- Thus, the random numbers are and corresponding names of the wind are,
Serial No. | Name |
01 | Arlene |
06 | Franklin |
12 | Lee |
13 | Maria |
14 | Nate |
18 | Stan |
19 | Unnamed |
27 | Epsilon |
The average maximum wind of above selected sample is calculated by using the following formula.
Thus, the sample mean is 75.
The population mean of maximum wind is,
Thus, the population mean is 87.5.
Comparison of means:
The sample mean of maximum wind speed is 75 and the population mean of maximum wind speed is 87.5.
That implies that the sample mean is smaller than the population mean. It is clear that there is some difference between the sample mean and the population mean.
Want to see more full solutions like this?
Chapter 14 Solutions
Bluman, Elementary Statistics: A Step By Step Approach, © 2015, 9e, Student Edition (reinforced Binding) (a/p Statistics)
- 13. A pharmaceutical company has developed a new drug for depression. There is a concern, however, that the drug also raises the blood pressure of its users. A researcher wants to conduct a test to validate this claim. Would the manager of the pharmaceutical company be more concerned about a Type I error or a Type II error? Explain.arrow_forwardFind the z score that corresponds to the given area 30% below z.arrow_forwardFind the following probability P(z<-.24)arrow_forward
- 3. Explain why the following statements are not correct. a. "With my methodological approach, I can reduce the Type I error with the given sample information without changing the Type II error." b. "I have already decided how much of the Type I error I am going to allow. A bigger sample will not change either the Type I or Type II error." C. "I can reduce the Type II error by making it difficult to reject the null hypothesis." d. "By making it easy to reject the null hypothesis, I am reducing the Type I error."arrow_forwardGiven the following sample data values: 7, 12, 15, 9, 15, 13, 12, 10, 18,12 Find the following: a) Σ x= b) x² = c) x = n d) Median = e) Midrange x = (Enter a whole number) (Enter a whole number) (use one decimal place accuracy) (use one decimal place accuracy) (use one decimal place accuracy) f) the range= g) the variance, s² (Enter a whole number) f) Standard Deviation, s = (use one decimal place accuracy) Use the formula s² ·Σx² -(x)² n(n-1) nΣ x²-(x)² 2 Use the formula s = n(n-1) (use one decimal place accuracy)arrow_forwardTable of hours of television watched per week: 11 15 24 34 36 22 20 30 12 32 24 36 42 36 42 26 37 39 48 35 26 29 27 81276 40 54 47 KARKE 31 35 42 75 35 46 36 42 65 28 54 65 28 23 28 23669 34 43 35 36 16 19 19 28212 Using the data above, construct a frequency table according the following classes: Number of Hours Frequency Relative Frequency 10-19 20-29 |30-39 40-49 50-59 60-69 70-79 80-89 From the frequency table above, find a) the lower class limits b) the upper class limits c) the class width d) the class boundaries Statistics 300 Frequency Tables and Pictures of Data, page 2 Using your frequency table, construct a frequency and a relative frequency histogram labeling both axes.arrow_forward
- Table of hours of television watched per week: 11 15 24 34 36 22 20 30 12 32 24 36 42 36 42 26 37 39 48 35 26 29 27 81276 40 54 47 KARKE 31 35 42 75 35 46 36 42 65 28 54 65 28 23 28 23669 34 43 35 36 16 19 19 28212 Using the data above, construct a frequency table according the following classes: Number of Hours Frequency Relative Frequency 10-19 20-29 |30-39 40-49 50-59 60-69 70-79 80-89 From the frequency table above, find a) the lower class limits b) the upper class limits c) the class width d) the class boundaries Statistics 300 Frequency Tables and Pictures of Data, page 2 Using your frequency table, construct a frequency and a relative frequency histogram labeling both axes.arrow_forwardA study was undertaken to compare respiratory responses of hypnotized and unhypnotized subjects. The following data represent total ventilation measured in liters of air per minute per square meter of body area for two independent (and randomly chosen) samples. Analyze these data using the appropriate non-parametric hypothesis test. Unhypnotized: 5.0 5.3 5.3 5.4 5.9 6.2 6.6 6.7 Hypnotized: 5.8 5.9 6.2 6.6 6.7 6.1 7.3 7.4arrow_forwardThe class will include a data exercise where students will be introduced to publicly available data sources. Students will gain experience in manipulating data from the web and applying it to understanding the economic and demographic conditions of regions in the U.S. Regions and topics of focus will be determined (by the student with instructor approval) prior to April. What data exercise can I do to fulfill this requirement? Please explain.arrow_forward
- Consider the ceocomp dataset of compensation information for the CEO’s of 100 U.S. companies. We wish to fit aregression model to assess the relationship between CEO compensation in thousands of dollars (includes salary andbonus, but not stock gains) and the following variates:AGE: The CEOs age, in yearsEDUCATN: The CEO’s education level (1 = no college degree; 2 = college/undergrad. degree; 3 = grad. degree)BACKGRD: Background type(1= banking/financial; 2 = sales/marketing; 3 = technical; 4 = legal; 5 = other)TENURE: Number of years employed by the firmEXPER: Number of years as the firm CEOSALES: Sales revenues, in millions of dollarsVAL: Market value of the CEO's stock, in natural logarithm unitsPCNTOWN: Percentage of firm's market value owned by the CEOPROF: Profits of the firm, before taxes, in millions of dollars1) Create a scatterplot matrix for this dataset. Briefly comment on the observed relationships between compensationand the other variates.Note that companies with negative…arrow_forward6 (Model Selection, Estimation and Prediction of GARCH) Consider the daily returns rt of General Electric Company stock (ticker: "GE") from "2021-01-01" to "2024-03-31", comprising a total of 813 daily returns. Using the "fGarch" package of R, outputs of fitting three GARCH models to the returns are given at the end of this question. Model 1 ARCH (1) with standard normal innovations; Model 2 Model 3 GARCH (1, 1) with Student-t innovations; GARCH (2, 2) with Student-t innovations; Based on the outputs, answer the following questions. (a) What can be inferred from the Standardized Residual Tests conducted on Model 1? (b) Which model do you recommend for prediction between Model 2 and Model 3? Why? (c) Write down the fitted model for the model that you recommended in Part (b). (d) Using the model recommended in Part (b), predict the conditional volatility in the next trading day, specifically trading day 814.arrow_forward4 (MLE of ARCH) Suppose rt follows ARCH(2) with E(rt) = 0, rt = ut, ut = στει, σε where {+} is a sequence of independent and identically distributed (iid) standard normal random variables. With observations r₁,...,, write down the log-likelihood function for the model esti- mation.arrow_forward
- MATLAB: An Introduction with ApplicationsStatisticsISBN:9781119256830Author:Amos GilatPublisher:John Wiley & Sons IncProbability and Statistics for Engineering and th...StatisticsISBN:9781305251809Author:Jay L. DevorePublisher:Cengage LearningStatistics for The Behavioral Sciences (MindTap C...StatisticsISBN:9781305504912Author:Frederick J Gravetter, Larry B. WallnauPublisher:Cengage Learning
- Elementary Statistics: Picturing the World (7th E...StatisticsISBN:9780134683416Author:Ron Larson, Betsy FarberPublisher:PEARSONThe Basic Practice of StatisticsStatisticsISBN:9781319042578Author:David S. Moore, William I. Notz, Michael A. FlignerPublisher:W. H. FreemanIntroduction to the Practice of StatisticsStatisticsISBN:9781319013387Author:David S. Moore, George P. McCabe, Bruce A. CraigPublisher:W. H. Freeman





