A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
Assessing and solving multicollinearity in sediment transport prediction models using principal component analysis
Predicting sediment transport in a river system is necessary when attempting to monitor river changes due to incoming particle deposition from upstream reaches. Previous studies have used various techniques such as probabilistic, regression, and equal mobility approaches to make predictions. However, issues associated with multicollinearity occur when using the regression approach. Significant correlations between the explanatory variables in a regression equation can increase the coefficient variance estimates, which make these estimates very sensitive to minor changes. Thus, we propose the use of principal component analysis as a solution to these issues. We observed that the variance inflation factor is less than 10, which indicates the nonexistence of issues associated with multicollinearity. Fitting components to the regression equation create a new prediction model for sediment transport, which improves upon previous equations using the same technique.
Assessing and solving multicollinearity in sediment transport prediction models using principal component analysis
Predicting sediment transport in a river system is necessary when attempting to monitor river changes due to incoming particle deposition from upstream reaches. Previous studies have used various techniques such as probabilistic, regression, and equal mobility approaches to make predictions. However, issues associated with multicollinearity occur when using the regression approach. Significant correlations between the explanatory variables in a regression equation can increase the coefficient variance estimates, which make these estimates very sensitive to minor changes. Thus, we propose the use of principal component analysis as a solution to these issues. We observed that the variance inflation factor is less than 10, which indicates the nonexistence of issues associated with multicollinearity. Fitting components to the regression equation create a new prediction model for sediment transport, which improves upon previous equations using the same technique.
Assessing and solving multicollinearity in sediment transport prediction models using principal component analysis
Sulaiman, Mohd Sofiyan (author) / Abood, Manal Mohsen (author) / Sinnakaudan, Shanker Kumar (author) / Shukor, Mohd Rizal (author) / You, Goh Qiu (author) / Chung, Xiang Zhi (author)
ISH Journal of Hydraulic Engineering ; 27 ; 343-353
2021-11-02
11 pages
Article (Journal)
Electronic Resource
Unknown
Causality, Measurement Error and Multicollinearity in Epidemiology
| Online Contents | 1996
ITERATIVE AGGREGATION METHOD FOR SOLVING PRINCIPAL COMPONENT ANALYSIS PROBLEMS
| DOAJ | 2017
Predicting Non-deposition Sediment Transport in Clean Pipes: Assessing Heuristic Models
| Springer Verlag | 2022
Numerical solving techniques for modeling sediment transport
| Tema Archive | 2000