Question

Which of the following is not a condition required for the regression procedure used in the exercises above to be statistically valid? Select all that apply.

A. The true errors must be normally-distributed.
B. The explanatory variable must be normally-distributed
C. The true relationship between the response and explanatory variables must be linear
D. The size of the sample must be sufficiently large, i.e., greater than n = 25
E. The response variable must be normally-distributed.

Answer 1

Answer: A, E, B, and C

Step-by-step explanation:

A regression procedure require all of the following

1. Linearity: The relationship between X and the mean of Y must always be linear. C fulfills this condition

2. Homoscedasticity: Sometimes known as the homogeneity of avariance is always same for any given value of X. A also fulfills this condition.

3. Normality: A fixed values of X, and Y is always normally distributed. B and E fulfills this condition.