- CFA Exams
- CFA Level I Exam
- Topic 1. Quantitative Methods
- Learning Module 1. Basics of Multiple Regression and Underlying Assumptions
- Subject 3. Assumptions Underlying Multiple Linear Regression
CFA Practice Question
Which of the following would not be an assumption underlying a multiple regression model?
B. There must be a linear relationship between the dependent variable and every independent variable.
C. The error term has a variance that is the same for all observations.
D. A linear relationship must also exist between two or more of the independent variables.
A. The error terms are normally distributed.
B. There must be a linear relationship between the dependent variable and every independent variable.
C. The error term has a variance that is the same for all observations.
D. A linear relationship must also exist between two or more of the independent variables.
Correct Answer: D
The fact is that there must be no perfect linear relationship among the independent variables. In other words, there should be no "multicollinearity" among the independent variables.
User Contributed Comments 3
| User | Comment |
|---|---|
| amamed213 | i think B is not an assumption. the modele don't require a linear relationship between the dependant variable and every independant variable. Can someone explain more ? |
| Adi8232 | If there is not a linear relationship between the dependent variable(Y) and ANY independent Variable(say X3) then why will you use that X3 in the equation, because you want value of Y and adding some non related will be useless (ex. Car mileage(Y) and number of candidates passing CFA L2(X)), will you use Y= a + bX? |
| Mikehuynh | Hi amanmed213, Just need to remember that we are talking about linear regression |