BUAD 310 Case 1 (Total 100 points) Due Thursday, December 1, 2011 in class In this case you will apply statistical techniques you learned in the...

BUAD 310 Case 1(Total 100 points)Due Thursday, December 1, 2011 in classIn this case you will apply statistical techniques you learned in the regression part of BUAD 310.Instructions:•This case project uses data from the file MagazineAds.MTW which you can download from Blackboard. •The entire case should be typed in word document and clearly written and submitted with supporting documents (Minitab printouts). The printouts can be directly copied and pasted into the word document. You need to include the relevant graphs and numerical output so we can cross-check your answers with the output.Magazine AdvertisingWhat factors influence the price of advertisements in magazines? Suppose you are part of a team of consultants hired by a retail clothing company wishing to place advertisements in at least one magazine. They are curious about what types of costs they can expect for magazines with different readership bases so they most effectively utilize their advertising budget. Your team has collected a data set of 57 consumer magazines and has found that the mean cost for a one-page advertisement is $48,641, but the standard deviation is $18,729. What number should be used to best estimate the advertising costs? Your team realizes that there may be many variables affecting cost of a one-page advertisement. You have augmented the original data set of 57 magazines by measuring more characteristics of the magazines and their audiences that may be useful in understanding the one-page advertisement costs better. The variables are the following :Magazine NameCost of a four-color, one-page adCirculation (projected, in thousands)Percent male among the predicted readershipMedian household income of readershipYour goal is to analyze the data with Minitab using Multiple Linear Regression methods and choose the best model to explain the differences in advertising costs between the different titles and then to predict what the retail clothing company should expect to pay for advertising in the different magazines. Answer the following questions:1.[12 pts] Examine the variables and their relationships to each other:a.First look at how each of the four variables behaves on its own by creating histograms of each. [Minitab: Graph → Histogram → Simple] Is there any apparent skewness in any of the graphs? Explain.b.Now explore the linear relationship between pagecost and each of the explanatory variables individually by constructing scatterplots of all three pairs. [Minitab: Graph → Scatterplot → Simple] Do you see any strong relationships? Are they linear? Explain your answer.2. [30 pts] Perform a Multiple Linear Regression analysis using all the explanatory variables AND perform a residual analysis using the graphs. [Minitab: Stat → Regression → Regression. After filling in the proper Response and Predictor variables, choose Graphs and check the box for Residuals versus fits.]a.Is this multiple regression model useful? Provide statistical evidence to support your answer and where appropriate use a significance level of 5%.b.What is the estimated regression equation?c.Examine each of the explanatory variables individually to determine which are contributing significantly to the model. Which independent variables would you recommend keeping in the model? (Use a significance level of 5%.) [Note: do not eliminate any variable(s) at this stage.]d.Evaluate the regression assumptions of linearity and homoscedasticity (constant variance of the error term) by assessing your residual plot. Be specific about your evaluation and describe any suggestions you have for remedying any problems. [Hint: for suggestions you may read questions 3 and 4 below.]e.Using this model with ALL the variables, provide a point estimate and an appropriate 95% interval to the retail clothing company for the amount that they should expect to pay for a full-page ad in a magazine with a projected audience of 4,000,000 readers, 60 percent of which are male, with a median income of $50,000. Include notation and units. Interpret these results.3.[18 pts] Often, when dealing with dependent variables that represent financial data (income, price, etc.), using the natural log of the dependent variable will help to alleviate problems that may be causing patterns in residuals/violations of the required conditions. Re-run the Multiple Regression analysis using the natural log of the page cost variable instead. [Minitab: Calc → Calculator, then enter “Lncost” in the Store Result in Variable. In Expression, enter “LN(pagecost)” and click OK. This will create a new variable that will be the natural log of the page cost variable.]. Re-run the Regression using this new variable as the dependent variable against all 3 independent variables, again creating residual plot for this model.a.Is this new multiple regression model useful? Provide statistical evidence to support your answer and where appropriate use a significance level of 5%. Does the new Regression model seem better that the previous ones? Why or why not?b.Examine each of the explanatory variables individually to determine which are contributing significantly to the new model. Use a significance level of 5%. Which explanatory variables would you recommend keeping in the new model? [Note: do not eliminate any variable(s) at this stage.] How does this compare to the results in question 2?c.Evaluate the regression assumptions of linearity and homoscedasticity by assessing your new residual plot. Be specific about your evaluation and describe any suggestions you have for remedying any problems. [Hint: for suggestions you may read question 4 below.]4.[24 pts] Since you have switched to using the natural log of the pagecost variable, you now need to re-create scatterplots using this as your dependent variable and each of the 3 independent variables on the x-axis (the result will be 3 separate scatterplots). The circulation variable has the most noticeable relationship to the natural log of pagecost. This is a logarithmic type of relationship; to transform this curved relationship into a linear kind, a natural log transformation needs to be applied to the circulation variable. Do the transformation in Minitab. [The steps are described in part 3. This will create a new variable that will be the transformed version of the original circulation variable.] Re-run the Regression and residual analysis using the new variable in place of the old one (leave the dependent variable as natural log of pagecost and leave the other two explanatory variables as they are).a.Has the usefulness of the model changed? Is this model better or worse than the previous model? Support your answer.b.Examine each of the independent variables individually to determine which are contributing significantly to the newest model. Use a significance level of 5%. Which variables would you recommend keeping in the model? How does this compare to the prior results?c.Evaluate the regression assumptions of linearity and homoscedasticity by assessing your new residual plot. Briefly comment on the new results.d.Finally, remove any variables that you deem to be insignificant at 10% level and re-run the model (remove the most insignificant variable at a time until all remaining explanatory variables in the model are significant at 10% level). Using this model, provide a point estimate and an appropriate 95% interval to the retail clothing company client for the amount that they should expect to pay on average if they were to place many ads in magazines with the following characteristics: projected circulation of 4,000,000 readers, 60 percent male, and median income of $50,000. Include notation and units. Interpret your results.EExecutive Summary: (1 page) [16 pts]You are given the task of summarizing your findings for the board of directors of the retail clothing company. Since they are not all very well-versed in regression techniques, you will need to explain things in easy-to-understand terms. Within the summary, explain which model and estimates you would recommend to best forecast the cost of one-page advertisements. Also, describe what this model indicates (very briefly) about the relationship between the pagecost and your chosen variables. If you feel your final model does not sufficiently explain pagecost, include your recommendations for improving it.