Notes for Top Assignment Statistics Expert:
1. Please answer ALL questions in American English spelling, 500 words total, open format.
2. Preferably, please utilize R – however, use whatever method necessary to answer questions.
These concepts can appear on the optional short-answer part of the tests. As part of this homework, answer the following questions, usually just several sentences that include the definition.
1. Identify the goals of preparing marketing research reports and presentations.
1. Describe the various components of a marketing research report.
1. Explain the four principles of an executive-ready report.
1. Discuss the use of oral presentations in marketing research reporting.
1. State the two goals of regression analysis.
1. How does multiple regression enhance the two primary goals of regression analysis?
1. How does b1 differ across the following regression models?
= b0 + b1X1 vs. = b0 + b1X1 + b2X2 ?
1. What is the purpose of model selection?
1. What are the criteria that a potential predictor variable should satisfy before added to a model?
1. What is a parsimonious model? Why is it desired?
1. Describe the p-value method for each slope coefficient as a strategy for model selection.
1. Describe the best predictor variable subsets method for each slope coefficient as a strategy for model selection.
This homework involves the analysis of a real-world survey project for the customers of a restaurant in Dallas, TX, SFG. The questions here are the same questions to answer for your project. The questions for this homework, and their co
esponding solutions, provide the template for the multiple regression analysis for your project and for the Final.
ing/data/SFGsfg.csv <-- copy & paste to web
Customers respond to the individual items with a 7-pt Likert format, from 1 to 7. Assess the customer's perception of the outcome variable Satisfaction (x22) with the following item:
How satisfied are you with the SFG?
Not Satisfied Very
At All Satisfied
1 2 3 4 5 6 7
What are the reasons that contribute to customer satisfaction? For the outcome variable of Satisfaction (named x22 in the data table), consider the follow three potential contributors:
· Is a fun place to eat (x13)
· Has an attractive interior (x17)
· Has excellent taste (x18)
The customer evaluates each item on the following Likert scale.
1 2 3 4 5 6 7
Analysis Question: To what extent do perceived Fun, Attractiveness and Tastiness together account for Overall Satisfaction of the restaurant dining experience at Restaurant SFG?
Do questions a through r and t, u, and y from the template. Use the following information for Questions e and f.
a. Identify the response variable and the predictor variable(s).
. Show the scatterplot matrix (just one scatterplot for a single predictor) and co
elation coefficients of the relationship of each of the variables in the model with each other. From only this visual information, develop some intuition for the subsequent analysis.
i. Relevance: Do the predictor variables relate to the target (response) variable? Explain.
ii. Uniqueness: [If multiple predictor variables] Could collinearity be a problem? Explain.
iii. Model Selection: [If multiple predictor variables] Given the co
elations, what is the most likely candidate for the final model? Explain.
c. Write the estimated regression model.
d. Specify and interpret the sample slope coefficient.
e. Manually calculate the fitted/predicted value for the given values of predictor variables X.
f. Manually calculate the associated residual. Interpret for the given values of predictor variables X and response variable y.
Hypothesis Test: Applied to the one specified predictor variable
g. Specify the null hypothesis and its alternative for the hypothesis test of the slope coefficient.
[answer with respect to the specifics of this analysis, e.g., not Predictor 1 but the actual name of each predictor in this specific analysis]
h. Show and label the calculation of how many (estimated) standard e
ors the estimated slope coefficient, b, is from the hypothesized population value.
[define the concept with the relevant numbers of this specific analysis, with or without a formula]
i. Include and apply the definition of the p-value with the relevant numbers for this specific analysis.
[include the relevant numbers in this specific analysis as an application of the general definition]
j. Specify the basis for the statistical decision for the hypothesis test and the resulting statistical conclusion for alpha=0.05.
[be specific with the numbers from this analysis as to the evaluation of the null hypothesis]
k. Hypothesis Test: Interpretation, as an executive summary you would report to management.
[applied to the relevant numbers of this specific analysis to generalize the results to the population, with no jargon like p-value or t-value or null hypothesis]
Confidence Interval: Applied to the one specified one predictor variable
l. Specify the value that the confidence interval estimates.
[do not provide the confidence interval, which is the estimate, not the value that it estimates]
m. Apply the definition of the 95% margin of e
or for its computation using the relevant numbers of this analysis with 2 approximating the t-cutoff.
[show the definition in words of the concept by applying the relevant numbers of this specific analysis, with or without a formula]
n. Show the computations of the 95% confidence interval illustrated with the specific numbers from this analysis.
[show the definition of the concept but apply the relevant numbers of this specific analysis, formula optional]
o. Confidence Interval: Interpretation, as an executive summary you would report to management.
[no jargon, which includes the phrase “slope coefficient”, nothing about hypothesis tests]
p. Demonstrate the consistency of the confidence interval and hypothesis test using the specific numbers for this analysis for both results.
[comparison includes the specifics of the numbers for this specific analysis for both inferential results]
q. Evaluate fit with the standard deviation of residuals.
. Evaluate fit with R2 and PRESS R2, including their comparison. Does this value indicate reasonable fit?
s. Show any potential outliers and explain why they are outliers.
Model Selection [if multiple predictor variables]
t. Consider all the predictor variables simultaneously. Based on the p-values of the slope coefficients, are any predictor variables much less useful for predicting the response variable (target)? Why or why not?
u. Any collinearity problems? Why or why not?
v. Based on this information and the best subset analysis, which model do you recommend? Why?
w. For the 95% prediction interval of [response variable y] for [the values of predictor variables X], show the interval including its calculation (can approximate with the t-cutoff of 2).
x. Interpret the prediction interval.
y. What decision do you recommend to management based on these results?
e. value of all three predictor variables: 4
f. value of the response variable: 5