1. The figure shows part of the full regression model (refer to Southwest Full Model if you want to recreate the output): Regression Table Coefficient Constant 12.69939153 COUPON 3.754890986 New 2.395531494 HI 0.008425792 S_INCOME 0.001206678 E_INCOME 0.001374273 S_POP 3.40095E-06 E_POP 4.36313E-06 DISTANCE 0.074986255 PAX 0.000870942 SW = Yes 40.96959691 GATE = Free 20.57923053 VACATION=Yes -35.644441 SLOT=Free 16.24476725 Which of the following statements are accurate and fully supported by the output? Hint: There are two key things to keep in mind when we make statements about "the world" based on our regression models: 1) the regression coefficients always describe the average effect of the independent variables on the dependent variable 2) holding everything else constant Vacation routes tend to be cheaper than other routes. If the gates and slots are free, that will lower fares on average, everything else held constant. The variables S_Pop and E_Pop are not important. 2. Regression Table Coefficient Constant 41.01629148 DISTANCE 0.077445525 SW=No 77.04166986 Interaction(DISTANCE,SW=No ) 0.012412868 The figure above shows regression output for the model: Fare=a + b1 * Distance + b2 * (SW=No) + b3 * Distance * (SW=No). What is the correct economical interpretation for b1 = 0.077? Select the best option. (Hint: To fully understand all coefficients in the model, the key is to write out a regression model for routes where SW=yes. Then write out the regression model for routes where SW=no. Refer back to the residency notes as needed. Simplify these two regression formulas, then the economic interpretation of b1 should become clear.) There is no economical interpretation. When you increase distance by one mile, fare increases on average by 7.7 cents; everything else held constant. 7.7 cents is the average increase in fare per additional mile on routes where Southwest is absent. 7.7 cents is the average increase in fare per additional mile on routes where Southwest is present. When an independent variable is added to a regression model, the R2 will increase. True False