eduzhai > Physical Sciences > Physics Sciences >

Choosing Appropriate Regression Model in the Presence of Multicolinearity

  • Save

... pages left unread,continue reading

Document pages: 10 pages

Abstract: This work is geared towards detecting and solving the problem of multicolinearityin regression analysis. As such, Variance Inflation Factor (VIF) and theCondition Index (CI) were used as measures of such detection. Ridge Regression(RR) and the Principal Component Regression (PCR) were the two other approachesused in modeling apart from the conventional simple linear regression. For thepurpose of comparing the two methods, simulated data were used. Our task is toascertain the effectiveness of each of the methods based on their respectivemean square errors. From the result, we found that Ridge Regression (RR) method is better than principal component regression when multicollinearity existsamong the predictors.

Please select stars to rate!

         

0 comments Sign in to leave a comment.

    Data loading, please wait...
×