90.2K

Verified Solution

Link Copied!

1. Tamson developed a model that used the actual price of a gallon of regular gasoline to predict motor fuel consumed per year. Based on the value of Durbin-Watson statistic in the following Minitab output, determine whether serial correlation is a problem. (10%)

Regression Analysis: Gallons versus Price

The regression equation is

Gallons = 87.8 - 14.7 Price

Predictor Coef SE Coef T P

Constant 87.802 2.119 41.43 0.000

Price -14.685 2.239 -6.56 0.000

S = 2.67384 R-Sq = 75.4% R-Sq(adj) = 73.7%

Analysis of Variance

Source DF SS MS F P

Regression 1 307.49 307.49 43.01 0.000

Residual Error 14 100.09 7.15

Total 15 407.58

Unusual Observations

Obs Price Gallons Fit SE Fit Residual St Resid

2 0.53 85.100 80.018 1.061 5.082 2.07R

R denotes an observation with a large standardized residual.

Durbin-Watson statistic = 1.94584

2. Tamson decides to add a variable representing the population of the United States to the model. Based on the output below, will you keep this new variable in this model? Explain. (10%)

Regression Analysis: Gallons versus Price, Pop.

The regression equation is

Gallons = 129 - 9.59 Price - 0.202 Pop.

Predictor Coef SE Coef T P VIF

Constant 129.29 16.58 7.80 0.000

Price -9.588 2.780 -3.45 0.004 2.1

16. -0.20157 0.08007 -2.52 0.026 2.1

S = 2.27505 R-Sq = 83.5% R-Sq(adj) = 81.0%

Analysis of Variance

Source DF SS MS F P

Regression 2 340.29 170.15 32.87 0.000

Residual Error 13 67.29 5.18

Total 15 407.58

Source DF Seq SS

Price 1 307.49

16. 1 32.81

Unusual Observations

Obs Price Gallons Fit SE Fit Residual St Resid

1 0.39 78.800 82.833 1.163 -4.033 -2.06R

R denotes an observation with a large standardized residual.

Durbin-Watson statistic = 2.37205

3. Is multicollinearity a problem in this case? If so, how to tackle this problem. (10%)

Answer & Explanation Solved by verified expert

See Answer