(b) Generate a 95% confidence interval for the slope in this regression. (c) Estimate the electricity usage for an apartment measuring 1,500 sq.ft. A single number estimate is sufficient, there is no need to calculate a prediction interval. (d) What percent of the variation of the electricity usage is explained by the variation of the size? (e) In another regression run, the square of the size variable has been included in addition to the size. The new column is called SIZESQ. The Minitab output is available in "Documents", under "Final exam, second Minitab output". Is the model including SIZESQ better than the one not including it? A yeso is not enough, give a reason. (f) The electricity usage of a particular 2,000 sq ft. apartment 1,820 kilowatt-hours. The 95% confidence interval Minitab created at SIZE=2,000 is (1, 712 ? 1, 796). Can you conclude that this apartment has an unusually high electricity usage?

ELECTRIC.MTW Regression Analysis: USAGE versus SIZE Regression Equation USAGE =903 + 0.3559 SIZE Coefficients Term Coef SE CoefT-Value P-Value VIF Constant 903 132 6.83 0.000 SIZE 0.3559 0.0548 6.50 0.000 1.00 Model Summary S R-sq R-sq(adj) R-sq(pred) 155.251 76.46% 74.65% 67.48% Analysis of Variance Source DF Adj SS Adj MS F-Value P-Value Regression 1 1017803 1017803 42.23 0.000 SIZE 1 1017803 1017803 42.23 0.000 Error 13 313338 24103 Total 14 1331140