6. Dummy variables review Aa Aa E A used car dealer believes a car's sale price can be predicted from the car's mileage and color. This Excel file contains the selling prices, mileages, and colors for a sample of 100 cars. The colors are coded as: 1. White 2. Silver 3. Other Run a regression of Price on Odometer and Color. The estimated slope coefficient on Color can be interpreted as: O When Color goes down by one unit, sale price decreases by 13.08 dollars. O When Color goes up by one unit, sale price increases by 13.08 dollars. 0 It does not have a sensible interpretation, because Color is a qualitative variable and is not properly coded. Create indicator (dummy) variables White and Silver, valued at 1 or 0, derived from the color information. Then run a regression on Odometer, White, and Silver. Which of the following are true? Check all that apply. Silver cars sell for, on average, 149.48 more than cars of other colors except white. For every additional mile on a car's odometer, the average decline in price is about 2.3 cents, holding other variables constant. Silver cars sell for, on average, 115.62 more than white cars. The fact that a car is silver has no statistical significance. The mileage variable, as represented by Odometer, is not statistically significant. Silver cars sell for, on average, 149.48 more than white cars. The fact that a car is white has no statistical significance. Silver cars sell for, on average, 115.62 less than white cars. At this stage, consider dropping any variables that are statistically not significant. Which variable, if any, should be dropped? Silver O None O Odometer O White Run a regression on the data, dropping that variable. Cars of the color identified in the final regression have an average price differential over cars of all other colors of 131.05 -13.1 115.6 149.5 33.9 Color Price 5054 5115 5410 5529 5507 5873 5303 5237 5383 5401 5595 5330 5806 5805 5317 5316 5870 5504 5333 5705 5150 5249 5775 5327 5192 5544 5054 5115 5410 5529 5507 5873 5303 5237 Odometer 34844 27379 47875 35648 42501 43803 43481 34279 41370 34966 41427 30241 32744 34470 37720 41350 24469 35781 48613 24188 38775 45563 28676 38231 36683 32517 39050 45251 34384 38383 32161 26561 33533 41849 WENN ww PNPWPNNWNNWPPNWPNWPPPNNPENN 5383 5286 5827 5483 5440 5215 5105 5685 5788 5208 5168 5128 5750 4965 5238 5763 5162 5486 5257 5228 5135 5267 5290 5387 5091 5667 5146 5765 5911 5532 5499 5204 5271 5007 36668 37495 25629 40099 31014 42233 37407 34356 30599 42485 38430 40452 26030 46296 34844 27379 47875 35648 42501 43803 43481 34279 41370 34966 41427 30241 47228 24464 21221 35521 28006 38079 42332 49223 NwNw w w w PNwNPPENPAW 2 5337 5066 5605 5637 5286 5820 5187 5435 5499 4787 5707 5457 5365 5160 5551 5568 5122 5854 5297 5148 5416 5733 5478 5690 5286 5745 5525 5424 5283 5259 5356 5133 33358 37819 35975 38085 35236 20962 45808 36183 34399 44330 32063 34641 31049 38636 36468 25745 39198 21535 37135 42581 33023 31644 35969 29051 38180 31494 31372 36238 34212 33190 39196 36392 NNNPWPNWPN w WNNW NANNWAND 6. Dummy variables review Aa Aa E A used car dealer believes a car's sale price can be predicted from the car's mileage and color. This Excel file contains the selling prices, mileages, and colors for a sample of 100 cars. The colors are coded as: 1. White 2. Silver 3. Other Run a regression of Price on Odometer and Color. The estimated slope coefficient on Color can be interpreted as: O When Color goes down by one unit, sale price decreases by 13.08 dollars. O When Color goes up by one unit, sale price increases by 13.08 dollars. 0 It does not have a sensible interpretation, because Color is a qualitative variable and is not properly coded. Create indicator (dummy) variables White and Silver, valued at 1 or 0, derived from the color information. Then run a regression on Odometer, White, and Silver. Which of the following are true? Check all that apply. Silver cars sell for, on average, 149.48 more than cars of other colors except white. For every additional mile on a car's odometer, the average decline in price is about 2.3 cents, holding other variables constant. Silver cars sell for, on average, 115.62 more than white cars. The fact that a car is silver has no statistical significance. The mileage variable, as represented by Odometer, is not statistically significant. Silver cars sell for, on average, 149.48 more than white cars. The fact that a car is white has no statistical significance. Silver cars sell for, on average, 115.62 less than white cars. At this stage, consider dropping any variables that are statistically not significant. Which variable, if any, should be dropped? Silver O None O Odometer O White Run a regression on the data, dropping that variable. Cars of the color identified in the final regression have an average price differential over cars of all other colors of 131.05 -13.1 115.6 149.5 33.9 Color Price 5054 5115 5410 5529 5507 5873 5303 5237 5383 5401 5595 5330 5806 5805 5317 5316 5870 5504 5333 5705 5150 5249 5775 5327 5192 5544 5054 5115 5410 5529 5507 5873 5303 5237 Odometer 34844 27379 47875 35648 42501 43803 43481 34279 41370 34966 41427 30241 32744 34470 37720 41350 24469 35781 48613 24188 38775 45563 28676 38231 36683 32517 39050 45251 34384 38383 32161 26561 33533 41849 WENN ww PNPWPNNWNNWPPNWPNWPPPNNPENN 5383 5286 5827 5483 5440 5215 5105 5685 5788 5208 5168 5128 5750 4965 5238 5763 5162 5486 5257 5228 5135 5267 5290 5387 5091 5667 5146 5765 5911 5532 5499 5204 5271 5007 36668 37495 25629 40099 31014 42233 37407 34356 30599 42485 38430 40452 26030 46296 34844 27379 47875 35648 42501 43803 43481 34279 41370 34966 41427 30241 47228 24464 21221 35521 28006 38079 42332 49223 NwNw w w w PNwNPPENPAW 2 5337 5066 5605 5637 5286 5820 5187 5435 5499 4787 5707 5457 5365 5160 5551 5568 5122 5854 5297 5148 5416 5733 5478 5690 5286 5745 5525 5424 5283 5259 5356 5133 33358 37819 35975 38085 35236 20962 45808 36183 34399 44330 32063 34641 31049 38636 36468 25745 39198 21535 37135 42581 33023 31644 35969 29051 38180 31494 31372 36238 34212 33190 39196 36392 NNNPWPNWPN w WNNW NANNWAND