A metalworking firm conducts an energy study using multiple regression methods. The dependent variable is y = energy consumption cost per day (in thousands of dollars), and the ­independent variables are x1 = tons of metal processed per day, x2 = average external temperature, x3 = rated wattage for machinery in use, and x4 = x1x2. The data are analyzed by Statistix. Selected output is shown here:

a. Write the estimated model.
b. Summarize the results of the various t tests.
c. Calculate a 95% confidence interval for the coefficient of METXTEMP.
d. What does the VIF column of the output indicate about collinearity problems?

Transcribed Image Text:

CORRELATINS PEARSON) ENERGY ETALHETXTEMP TEMe 0.612 METAL 0.1094 TEMP 0-4007 -0.0606 0.9931 0.5775 02239 0.3630 0.3529 UNNEIGHTED LEAST SOUARES LINEAR REGRESSION OF ENERGY PREDICTOR VARTABLES COEFFICIENT S BRROR STULENT'S T 7.20439 1-36291 0 30588 0.01024 -0.00277 17-5322 0:92438 1.62104 0:00473 0.07722 0. 41 1.47 D.19 2-16 -0.04 0 6855 0.1559 3.8 0.6522 250.0 1.5 0.9717 246.4 METAL TEN 0.0427 R-SQUARED ADJUSTED R-SOUARED ,5963 STANDARD DEVIATIN 6.51555 2.55255 0.6636RESID. MEAN SQUARE (MSE) SOURCE DF REGRESSION 4 257.0484.2622 9.RE D.0001 RESIDUAL TOTAL 20 130 311 .51555 24 397.360 CASES INCLULED 25MISSING CASES