Question;Even though the housing market has deteriorated in recent months, it is important to determine correct valuations of homes. A real estate appraiser will look at comparables, quite often, as the initial step in determining the value of a home. Besides comparables, most appraisers will look at various features of homes in establishing an appraised value. For example, a hot tub might increase an asking price of a home by $5,000 while a home that is in need of a paint job might see a decrease in valuation by $5,000. A useful tool that appraisers should use is regression analysis. With this in mind, I have a dataset containing 76 home sales in a western U.S. community.Although this problem is not directly connected to cost accounting, the concept of regression analysis still applies to this problem. People are always trying to predict or estimate some value based on some known parameters. The object of regression analysis is to try to develop an equation that is strong in predicting some unknown factor (in this case, sales price).Answer the following questions (each question should be a separate worksheet tab labeled as q1, q2, etc.) (point values for each component are noted):1. (12) For each of the six potential sales price drivers (sq footage, baths, bedrooms, year built, car garage size, and fenced (yes=1, no = 0)) do each of the following:a. Prepare a scatter plot. The y axes should be sales price, and the x axes will be one of the potential drivers.i. Please title each axis so I know which sales price driver is being included in the chart.ii. Include a trend line for each plot along with the equation and R2.iii. Please make separate charts for each driver (no more than 2 per tab ?charts1, charts2, etc.).iv. Be sure to properly label the tabs so that I can find the charts in your Excel file.2. (12) Using the regression feature within data analysis, find the:i. slope,ii. R2, andiii. intercept for each sales price driver.3. (10) Based solely on the scatterplot appearance, without reference to the numbers you developed with the regression analysis, comment on the appropriateness of each potential sales price driver.4. (10) Based on the scatterplots, the regression information, and the information in the background materials, choose the best sales price driver. Explain your choice, including a discussion of why other sales price drivers were not chosen.5. (10) What is the regression equation that you found for the best sales price driver?6. (10) Explain what this ?best? regression equation tells you (specifically).7. (10) What is the ?best? multiple regression equation for sales price? You might have to create several equations with a combination of variables to find this equation (or, there might not be any useful sales price drivers from this dataset).8. (8) Assume that the following is the best multiple regression equation (it isn?t!).Estimated sales price = 41,005 +123.55(sq ft) +2,974(baths) + 957.65(fenced)Using the following information, what is the expected sales price:House features:Square feet = 1950Baths = 1.5Bedrooms = 3Garage = 2Fenced = noYear built = 19609. (8) Assume that a good estimator for sales price is garage size. The equation for this sales price driver is:Estimated sales price = 241,926.70 + 28,016.98(garage)Based on this equation, the residual for the first home ($388,000) is $146,073. Explain what this means?10. (10) List three other variables that you might suggest for predicting sales price of a home. How would you go about collecting the data for each variable (be specific)?
Paper#57338 | Written in 18-Jul-2015Price : $45