Segmentation Analysis $(120$ points$)$: $1.$ Describe the context of this case. That is$,$ what is the purpose for conducting this segmentation analysis? $(10$ points$)2.$ What are the primary segmentation variables used in this analysis? $($refer to the case Appendix$)(10$ points$)3.$ How many segments emerged from the data? Using the Segment Description data $($i$.$e$.,$ segment size and segment description charts$),$ to summarize each segment based on their importance ratings relative to the population. $(20$ points$)4.$ Take a look at the segment space. While this might be a little challenging, tell me what you can learn about the segments from their positioning relative to the two components $($x and y axis$).$ In your answer, be sure to infer what factors x and y represent. $(20$ points$)5.$ Take a look at the descriptor, table in the discriminant analysis section. Do the descriptors chosen for this model discriminate well? Why or why not? $(10$ points$)6.$ From the discriminant analysis, it appears that there are a few descriptors that offer some discrimination among segments. Describe the segments based on this analysis $($the color chart is quite helpful and be sure to go back to the case to see how these measures are coded$).(10$ points$)7.$ The discriminant model section shows the set of predictive models used to assign observations $($people$)$ to clusters using a multinomial logit model. In multinomial logit$,$ there are n$-1$ equations where n equals the number of clusters. A predicted probability is calculated for each person, for each segment. Observations are then assigned to a cluster based on the cluster for which they received the highest probability. The confusion matrix is created by showing the number of correct$/$incorrect assignments of the discriminant analysis compared to that of the cluster analysis. So$,$ the cluster analysis provides the actual assignment, and the discriminant analysis provides the predicted assignment. Overall, how well does this model perform in classifying observations? $(30$ points$)8.$ Discuss the model quality for the $3-$cluster solution. $(10$ points$)$ Positioning Analysis $(70$ points$)$: $1.$ A three$-$dimension solution emerges as opposed to a two$-$dimension solution. Explain the solution. $(20$ points$)2.$ Skip down to the summary section of the Attributes, section and describe the dimensions by summarizing the information in the chart. $(10$ points$)3.$ Characterize the perception of Kirin, specifically, using its position on the three maps. $(20$ points$)4.$ Do the maps offer any evidence that changing the price of Kirin would be advantageous? $(20$ points$)$Conjoint Analysis $(110$ points$)$:$1.$ List the product attribute and attribute levels used in the conjoint analysis. $(10$ points$)2.$ Describe the task respondents completed. That is$,$ describe the method by which the data were collected. $(10$ points$)3.$ Take a look at the Report Preference Partworths, output. For each attribute, indicate which level resulted in the highest average preference partworth. $(10$ points$)4.$ Based on the Simulations with Existing Products, how much market share does Kirin hold? $(10$ points$)5.$ Now, look at the Simulation with New Products. Describe the new Kirin. How much market share will it gain if introduced? $(30$ points$)6.$ Based on these simulations, how would you recommend Kirin modify the new formulation to meet consumer preferences optimally? $(40$ points$)$