Please answer the following 4 +1 questions. The first four questions of 25 points each are post Quiz1 material and the last question is a bonus question of 15 points. A bonus question is one where if you get it right, it adds to your total, but if you get wrong wont hurt you. You have 3 hours to complete the quiz. Ideally you should not spend more than 30 minutes on any question. Upload your word document with relevant output to the Assignment folder titled Quiz2. Note that you should include as much explanation as possible to your answers and include all relevant output within your Word document. Additionally, please include your entire XLSTAT/Excel output separately. This is for me to cross check information between your document and output in case of doubt. 1a. In the dataset (Quiz2_softdrink), you have the various attributes and levels on which consumers evaluate soft drinks. Please indicate how many profiles are possible. Then use Designs for Conjoint to indicate whether 7 or 8 profiles can be used to survey target consumers for their preference ranking of the profiles. If you find that either 7 or 8 profiles would work, would you still consider asking respondents to rank 10 profiles, and if so, why? 1b. You are given the IceCream dataset (Quiz2_icecream.xls). Assume that you are interested in launching a new brand with the following specifications: Vanilla, Homemade Waffle, Low fat, and Not Organic. You can assume that your main competitor is as in the table below. Please simulate the market share of this new brand when it competes with the competing brand (Profile 1) shown below. Profile1 Chocolate Pint No low fat Not organic 2a. A company has 1000 customers when it starts operations and has a churn rate of 25%. In 5 years, what percentage of its original customers will the company retain? Now assume that the company makes 1 dollar profit per customer and that customers stay with the company for 5 years. How much would the company be willing to spend to attract such customers who stay for 5 years? You may assume that the inflation rate is 5%. 2b. How sensitive are your results (about much you would be willing to spend) to your assumptions about retention rate and discount rate?. Please display the results for retention rates of 90 %, 75%, and 50% respectively and discount rate is 1%, 5%, and 10% using the Excel tool. 3a. What advantage does an experiment have over observation? Then, please specify the conditions that an experiment needs to satisfy in order for one to be able to make causal claims. 3b. Nintendo has a new video game for release. However, they are not sure what effect price and advertising have on sales. Hence, they have released their product in a test market by manipulating price and advertising to three levels of high/medium/low. The results of the test are in Quiz2_videogames. Please analyze this dataset and indicate what effect price and advertising have on the sales of video games, including any interaction effects. Make sure that you interpret the results properly keeping the base level to be Low in all instances. 4a. The following dataset (Quiz2_RFM.xls) relates to a direct marketer who has tracked how recently (R), frequently (F), and how much a person has bought from their online store (M). A survey of a random sample of their customer base was taken and their interest in a new item was taken (1=will buy new item, 0 will not buy item). The variables of R, F, and M in the dataset (RFM.xls) are on a 1-5 point scale where a 5 on R indicates a most recent purchase, a 5 on F indicates many purchases by the individual, and a 5 on M indicates a heavy spender. . Which variable is the best predictor of consumer interest in the new item? Also, what would the effect be of a unit increase in the variable that is the best predictor of item purchase?. 4b. Please explain the terms sensitivity and specificity mean with respect to classification accuracy? 5a. A company has surveyed 605 of its customers and asked them several questions as it pertains to their loyalty (Quiz2_loyaltyCorrectFormat.xls). The company has hired a recent USF graduate who has been given the task to explore the dataset to find out what factors influence the loyalty to the company. Please indicate which of the variables can be used to predict loyalty. Note that the variables are all ratings on a 7-point scale where 7 is the best and 1 is the worst. Only gender is coded as a qualitative variable (you may consider 1 to be male and 2 to be female). The company strongly believes that gender has a direct effect on loyalty and also influences genders effect on quality on loyalty as it believes that the effect of quality varies by gender. Kindly build a model that incorporates the effect of relevant variables that can predict loyalty as well as answer the questions posed by the companys belief in the effect of gender. Make sure to explain which variables significantly influence loyalty and explain/interpret the nature of the effect. Remember to specifically address whether the company is right or wrong in its assumptions about the role of gender on loyalty. 5b. How elastic is loyalty with respect to quality perceptions?. Make sure to show the steps of your calculation and interpret the result.
