20180711_Determination of the optimal parameters in regression models for the prediction...



1. Title

Determination of the optimal parameters in regression models for the prediction of chlorophyll-a: A case study of the Yeongsan Reservoir, Korea



2. Summary

  • Chl-a is used as the main dependent variable because of its simple, direct, and reasonable indicator for measuring phytoplankton biomass
  • The existence of correlation among explanatory variables (“collinearity”), causes the poor accuracy of prediction model
  • Main Topic
    • The backward stepwise method of MLR is useful for solving the collinearity problem?
    • Can PCR be used as an alternative approach for resolving the collinearity problem?
    • What is the best regression model with the least uncertainty?


3. Application

  • Complexity and noise in the original data were removed by PCR
  • F-overall-number of explanatory variables (RFN) curve, the proposed methodology selected four regression models as a basis(candidates) for further comparison of expanded uncertainties. 



4. Contact

Sung Ho Shin (Ph.D. program)

Environmental Systems Engineering Lab.

School of Earth Sciences and Environmental Engineering

Gwangju Institute of Science and Technology

1 Oryong-dong Buk-gu Gwangju, 500-712, Korea

 Phone : +82-10-6634-8614

E-mail : hogili89@gist.ac.kr




좀 더 세부적인 내용의 리뷰는 파일로 첨부하였습니다.

첨부 (1)
1.61MB / Download 2