Academic Integrity: tutoring, explanations, and feedback — we don’t complete graded work or submit on a student’s behalf.

D Question 10 In general, what should you do if you think multicolliearity is a

ID: 3061069 • Letter: D

Question

D Question 10 In general, what should you do if you think multicolliearity is a problem? O Check the correlation coefficients to make sure they are less than .3 across the X variables Throw out the X variables and see if the adj R2 has increased Throw out the X variables with VIF -8, one at time and make sure that P-values have decreased, choose the model with the highest adj. R2. After checking each X variable, include all of them in the final regression. Question 11 50 pts In general, why do we need to test the regression for curvilinearity? O We need to make sure the variances of the subpopulations of Y are all equal o We need to check to make sure that we put a curve in the regression line to fit the data, if needed O We need to check to make sure that there is not mismeasured X variables o We need to check to make sure there is not inflated standard errors (high VIF)

Explanation / Answer

Please post 1 question per post, as per forum rules.

10. C is the best option when faced with multicollinearity

When there is multicollinearity then VIF increases, blows up and and R2 also increases without reasons, due to overfitting.

To, avoid this we remove variables with high VIF and check for increase in Adj. Rsquare