Module 2.8 Multicollinearity

Question 1

refers to condition when two or more of independent variables or linear combinations of independent variables in a multiple regression are HIGHLY correlated with each other

Answer 1

multicollinearity

Question 2

distorts the standard error of estimate and the coefficient standard errors, leading to problems when conducting t tests

Answer 2

multicollinearity

Question 3

standard errors of slop coefficients are artificially INFLATED

Answer 3

multicollinearity

Question 4

what type of error is more common when multicollinearity exists

Answer 4

type 2 error (false negative)

Question 5

most common way to detect multicollinearity

Answer 5

1. t tests indicate NONE of coefficients are different than 0
2. f test is significant
3. R^2 is high

Question 6

suggests possibility of multicollinearity

Answer 6

high correlation amongst independent variables

Question 7

does not necessarily indicate multicollinearity is NOT present

Answer 7

low correlation among independent variables

Question 8

most common method to correct for multicollinearity

Answer 8

omit one or more of the correlated independent variables

Question 9

selection of explanatory/independent variables to be included in the regression and the transformations, fi any

Answer 9

regression model specification