# Question: An alternative to the regression formula y

An alternative to the regression formula μy = α + βx expresses each y value, rather than the mean of the y values, in terms of x. This approach models an observation on y as

y = mean + error = α + βx + ɛ,

Where the mean μy = α + βx and the error = ɛ.

The error term denoted by ɛ (the Greek letter epsilon) represents the deviation of the observation from the mean, that is, ɛ = error = y - mean.

a. If an observation has ɛ > 0, explain why the observation falls above the mean.

b. What does ɛ equal when the observation falls exactly at the mean? The ɛ term represents the error that results from using the mean value (α + βx) of y at a certain value of x for the prediction of the individual observation on y.

c. For the sample data and their prediction equation = α + βx, explain why an analogous equation to the population equation y = α + βx + ɛ, is y = a + bx + e, where e is the residual, e = y - . (The residual e estimates ɛ. We can interpret e as a sample residual and ɛ as a population residual.)

d. Explain why it does not make sense to use the simpler model, y = α + βx, which does not have the error term.

y = mean + error = α + βx + ɛ,

Where the mean μy = α + βx and the error = ɛ.

The error term denoted by ɛ (the Greek letter epsilon) represents the deviation of the observation from the mean, that is, ɛ = error = y - mean.

a. If an observation has ɛ > 0, explain why the observation falls above the mean.

b. What does ɛ equal when the observation falls exactly at the mean? The ɛ term represents the error that results from using the mean value (α + βx) of y at a certain value of x for the prediction of the individual observation on y.

c. For the sample data and their prediction equation = α + βx, explain why an analogous equation to the population equation y = α + βx + ɛ, is y = a + bx + e, where e is the residual, e = y - . (The residual e estimates ɛ. We can interpret e as a sample residual and ɛ as a population residual.)

d. Explain why it does not make sense to use the simpler model, y = α + βx, which does not have the error term.

**View Solution:**## Answer to relevant Questions

You invest $1000 at 6% compound interest a year. How long does it take until your investment is worth $2000? a. Based on what you know about exponential regression, explain why the answer is the value of x for which ...For the high school female athlete strength study, the output shows a correlation matrix for height, weight, body mass index (BMI), and percentage of body fat (BF%). a. Which pair of variables has the (i) Strongest ...Refer to the association you investigated in Exercise 12.8 between skipping class and college GPA. Using software or a calculator with the data file you constructed for that exercise, a. Find the mean and standard deviation ...Refer to the High School Female Athlete and Male Athlete Strength data files on the text CD. a. Find the correlation between number of 60-pound bench presses before fatigue and bench press maximum for females and between ...Suppose the regression line μy = -10,000 + 1000x models the relationship for the population of working adults in Canada between x = age and the mean of y = annual income (in Canadian dollars). The conditional distribution ...Post your question