· Page length: Three page maximum.
· The project involves:
· Collecting data suitable for an analysis of positioning on a minimum of 15 to 20 objects.
· Product-service price: this will serve as the dependent variable in your regression model
· Two other product attributes that will serve as independent variables in your regression model
· Creating a positioning scatterplot, examining and reporting on multiple regression model results, and providing an executive summary analyzing the situation
· Comments on selecting objects:
· To reduce extraneous variance and increase the relevancy of your results, you need to pay attention to the conditions under which the data is collected. For example, the objects must be in the same product category. Notice that in the sedan example, light trucks and SUVs were not included. In the hotel example, apartment and condominium rental units were not included. In addition, prices were collected for one specific day.
· If necessary, you should control for location and time: In the hotel example, hotels were restricted to downtown hotels and on a specific day.
· Please feel free to replicate in part the sedan and hotel examples, but with some twists. Examples:
· An analysis of Los Angeles hotels. You would model price as the dependent variable. One regression independent variable might be average review out of five. The other might be whether the hotel is located at the airport or in the downtown city center. This would allow you to understand the average price difference across locations. If this option is chosen, you should consider modeling 20 objects, with 10 in each location. On the other hand, you could model just one location and select a different attribute such as the presence of a swimming pool.
· An analysis of light trucks modeling horsepower and vehicle size as independent variables. You could model size as wheelbase length. You could engage in the same analysis, but for SUVs.
· Report format:
· Page 1:
· Project title and team
· Executive summary
· Pages 2 and 3: Charts, model, process notes, and additional technical details
· Provide a scatterplot using price on the vertical axis and one of your two independent variables on the horizontal axis.
· Use effective labeling and label your objects in the scatterplot
· Note that charting all three variables at once is challenging and is therefore not required. Please select just one of your independent variables and plot just that one.
· Data file (Excel): the actual data itself
· Variable descriptions. For example, price in the sedan example was defined as “Price was taken from Kelly’s Blue using the base automatic model MSRP.”
· Report the results of your regression model. The following template may be used (fictitious example):
Model r-square = .75
Model F = 12.278; p-value < or=""> .10
Regression model:
Predicted value of Y = Intercept + Slope estimate 1 x Variable 1 + slope estimate 2 x Variable 2
Regression model: Y = 2.79 + $8,000 x Sedan Horsepower + $8,200 x Sedan Size
Sedan Horsepower p-value < or=""> .10
Sedan Size p-value < or=""> .10