Constructing partial least squares model in the presence of missing data
Virtually all methods of data analysis are plagued by problems with missing data, and partial least squares are no exception.This work contributes to knowledge by critically examining ways of handling missing data in the estimation of Partial Least Squares Models. The experiments are performed using...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Published: |
American Scientific Publishers
2015
|
| Subjects: | |
| Online Access: | http://repo.uum.edu.my/16642/ http://doi.org/10.1166/asl.2015.6120 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Virtually all methods of data analysis are plagued by problems with missing data, and partial least squares are no exception.This work contributes to knowledge by critically examining ways of handling missing data in the estimation of Partial Least Squares Models. The experiments are performed using real world customer satisfaction dataset.The results indicate that Multiple Imputation performs better than the other methods for all percentages of missing data. Another unique contribution is found when comparing the results before and after the Neural Network post-processing procedure.This improvement in accuracy is resulted from the neural network’s ability to derive meaning from the imputed data set found by the statistical methods. |
|---|