In a regression setting, which, if any, of the following is not an assumption about the random error ?
A) The errors are linearly correlated.
B) The errors are normally distributed with a mean of 0 and a common variance.
C) The errors are independent in a probabilistic sense.
D) All of these are assumptions about the random error.
E) None of these.

Question

In a regression setting, which, if any, of the following is not an assumption about the random error  ?
A) The errors are linearly correlated. 
B) The errors are normally distributed with a mean of 0 and a common variance. 
C) The errors are independent in a probabilistic sense. 
D) All of these are assumptions about the random error. 
E) None of these.

Quizplus · Accepted Answer

The assumption that the errors are linearly correlated is not a necessary assumption in a regression setting. However, it is important to check for any patterns or trends in the residuals to ensure there is no violation of this assumption. The other three options (normal distribution, zero mean, and independence) are all important assumptions that are necessary for valid inference and interpretation in a regression analysis.

In a Regression Setting, Which, If Any, of the Following