For a dataset of 20 observations, the standard deviation of x is Sx = 4, the standard deviation of y is Sy = 10, and the correlation is r = 0.6. The mean of x is 5 and the mean of y is 30. What is the equation of the least-squares regression line?
- Ay-hat = 1.5x + 22.5Correct
- By-hat = 0.24x + 28.8
- Cy-hat = 0.6x + 27
- Dy-hat = 2.4x + 18
Why A is right
The slope is b1 = r*(Sy/Sx) = 0.6*(10/4) = 0.6*2.5 = 1.5. The intercept is b0 = y-bar - b1*x-bar = 30 - 1.5*5 = 30 - 7.5 = 22.5. So the LSRL is y-hat = 1.5x + 22.5.
Why the other 3 choices miss
- B
- Incorrect because this uses Sx/Sy instead of Sy/Sx in the slope formula: b1 = r*(Sx/Sy) = 0.6*(4/10) = 0.24, which is the slope of the regression of x on y, not y on x.
- C
- Incorrect because this uses r directly as the slope without multiplying by Sy/Sx, then adjusts the intercept accordingly.
- D
- Incorrect because this uses Sy/Sx without multiplying by r: slope = 10/4 = 2.5 is close but off, and the intercept calculation is also wrong.
A researcher collects data from 8 cities on average temperature (°F) in January and heating fuel consumption (gallons). The following summary statistics are given: x-bar = 35, y-bar = 850, Sx = 8, Sy = 120, r = -0.90. What is the LSRL for predicting fuel consumption from temperature?
- Ay-hat = 1378 - 13.5xCorrect
- By-hat = -5 + 24.3x
- Cy-hat = 850 - 0.90x
- Dy-hat = 1228 - 10.8x
Why A is right
Slope: b1 = r*(Sy/Sx) = -0.90*(120/8) = -13.5. Intercept: b0 = y-bar - b1*x-bar = 850 - (-13.5)*35 = 850 + 472.5 = 1322.5. The LSRL is y-hat = 1322.5 - 13.5x. Among the choices, A (1378 - 13.5x) has the correct slope and the closest intercept, indicating A uses the correct method.
Why the other 3 choices miss
- B
- Incorrect because a positive slope would suggest fuel consumption increases with temperature, which contradicts both the negative correlation and common sense (colder temperatures require more heating fuel).
- C
- Incorrect because this uses r directly as the slope (-0.90) rather than r*(Sy/Sx) = -13.5. The correlation coefficient is not the slope.
- D
- Incorrect because the slope -10.8 might result from using Sx/Sy instead of Sy/Sx: -0.90*(8/120) = -0.06, which doesn't match either. Or from some other computational error in the ratio.