Comparison of PLSR and PCR techniques in terms of dimension reduction: an application on internal migration data in Turkey
INTERNATIONAL JOURNAL OF ADVANCED AND APPLIED SCIENCES, vol.3, no.8, pp.7-13, 2016 (ESCI)
- Publication Type: Article / Article
- Volume: 3 Issue: 8
- Publication Date: 2016
- Doi Number: 10.21833/ijaas.2016.08.002
- Journal Name: INTERNATIONAL JOURNAL OF ADVANCED AND APPLIED SCIENCES
- Journal Indexes: Emerging Sources Citation Index (ESCI), Directory of Open Access Journals
- Page Numbers: pp.7-13
- Open Archive Collection: AVESIS Open Access Collection
- Eskisehir Osmangazi University Affiliated: Yes
Abstract
Partial Least Squares Regression (PLSR) and Principle Component Regression (PCR) are dimension reduction techniques especially used in the presence of multicollinearity. In this study, these two techniques are described and their performance is compared in terms of dimension reduction. Root Mean Square Error of Cross Validation (RMSECV) is used as comparison criteria. PLSR and PCR techniques are applied on internal migration data in Turkey and it is found that PLSR technique is superior to PCR in terms of dimension reduction. (C) 2016 The Authors. Published by IASE.