Consider a data set of prices of n = 202 new cars in 2015, where several manufac

Question

Consider a data set of prices of n = 202 new cars in 2015, where several manufacturers were selected and the car type is compact, coupe, sedan, or wagon. The price is MSRP (manufacturer suggested retail price). The aim is to find a prediction equation of MSRP (in thousands of dollars) based on explanatory variables: Displacement (Disp), Horsepower (Hp), Tanksize (Tank), and Brand Category (4 categories labelled as G1=economical, G2=upscale, G3=luxury, Porsche). Two models, respectively with MSRP and In(MSRP) as response variables, are fitted. Some outputs and residual plots from R are shown below. What is a noticeable pattern in one of the residual plots? non-normality sinusoidal quadratic heteroscedasticity Based on the above, which of the two models is better? Give at least 2 reasons for your choice. Model 2 Model 1 What is an approximate 95% confidence interval for the coefficient of BrandG1 in model 2? Lower limit is Upper limit is. Interpret beta_BrandG1^^and the interval in (c) What are in their corresponding rows of the X matrix (which has 7 columns), for model y~Brand+Disp+Hp+Tank? Assume the columns of X are for intercept, brandG1, brandG2, brandG3, Disp, Hp, Tank: X_1, 2 =, X_1, 4 =, X_1, 6 =, X_2, 1 =, X_2, 3 =, X_2, 5 = .

Explanation / Answer

The value you need to fetch from the data source

X1,2 = brandG2 in row 1

X1,4 = Disp in row 1

X1,6 = Tank in row 1

X2,1 = brandG1 in row 2

X2,3 =brandG3 in row 2

X2,5 = Hp in row 2

You can find these values in the data source

Related Questions

Navigate

• Browse (All)
• Browse C
• Subjects

---

---

Integrity-first tutoring: explanations and feedback only — we do not complete graded work. Learn more.