Advances and Applications in Statistics
Volume 25, Issue 2, Pages 109 - 113
(December 2011)
|
|
RESIDUAL BOOTSTRAP METHOD: AN APPROACH TO MODEL VALIDATION
T. O. Olatayo, A. I. Oredein and A. A. Akomolafe
|
Abstract: Model validation is an important step in the modeling process and helps in assessing the reliability of models before they can be used in decision making. We used residual bootstrap method to study regression model validation process. Various validation statistic such as the standard error (SE), mean square error (MSE), and coefficient of determination were used as criteria for selecting the best model. The values of these statistics revealed that the residual bootstrap approach gives better estimates, which reduces the risk of over fitted models. |
Keywords and phrases: residual bootstrap, model validation, standard error, mean square error, coefficient of determinations. |
|
Number of Downloads: 379 | Number of Views: 1896 |
|