Adaptive Regression for Modeling Nonlinear Relationships by George J. Knafl & Kai Ding

Author:George J. Knafl & Kai Ding
Language: eng
Format: epub
Publisher: Springer International Publishing, Cham

The adaptive unit dispersions model for merclevel in terms of weight, length, river, and GCs based on LCV scores is described in Sect. 8.11. It has the largest LCV score of 0.67602 for prior models of merclevel and PCDP score 0.66272 (112/169 correctly predicted merclevel values). The associated adaptive model based on PCDP scores has PCDP score 0.62722 (106/169 correctly predicted merclevel values) and LCV score 0.64885. The PD in the LCV scores for the PCDP-based model is substantial at 4.02 %. Once again, the LCV-based model outperforms the PCDP-based model on the basis of both PCDP and LCV scores. These results indicate that PCDP scores can generate distinctly inferior adaptive models for polytomous outcomes. Since this also is the case for dichotomous outcomes (Sect. 8.12.2), the use of PCDP scores is not recommended for adaptive modeling.

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.