|
Comparing Multiple Regression, Principal Componant Analysis, Partial Least Square Regression and Ridge Regression in Predicting Rangeland Biomass |
|---|---|
| รหัสดีโอไอ | |
| Creator | Ali Goharnejad, Azin Zarei, and Pejman Tahmasebi |
| Title | Comparing Multiple Regression, Principal Componant Analysis, Partial Least Square Regression and Ridge Regression in Predicting Rangeland Biomass |
| Publisher | คณะสิ่งแวดล้อมและทรัพยากรศาสตร์ |
| Publication Year | 2557 |
| Journal Title | Environment and Natural Resources Journal |
| Journal Vol. | 12 |
| Journal No. | 1 |
| Page no. | 1 |
| Keyword | Biomass, Multiple regressions, PCA ordination, Ridge regression, Partial Least Squares (PLS), NDVI index |
| ISSN | 1686-5459 |
| Abstract | In this paper, the prediction of rangeland biomass using different methods including Multiple regression, Principal Component Analysis, Partial Least Square regression and Ridge regression were compared. For this goal, environmental factors such as elevation (m), rainfall (mm), slope (?), caco3 (?), Sand (?), Sand (?), Clay (?), pH, EC (ds.m-1), Saturation (SP (?)) were used to determine a relationship between environmental factors and the forage yield. The results showed that PLS, and ridge regressions were among the best models to predict rangeland biomass followed by multiple regressions and Principal component Analysis. PLS and ridge regression had a higher predicted accuracy than other calculation methods. It was shown that, the Partial Least Square regression values to R, RMSE and MAE were 0.83, 34.9 and 26.23, respectively. |
