Universiti Teknologi Malaysia Institutional Repository

Modeling oil palm yield using multiple linear regression and robust m-regression

Yacob, A. Rahim and Ismail, Zuhaimy and Khamis, Azme (2005) Modeling oil palm yield using multiple linear regression and robust m-regression. In: Annual Fundamental Science Seminar 2005 (AFSS2005),Johor, 2005, Johor.

[img]
Preview
PDF (Abstract)
63Kb

Official URL: http://dx.doi.org/10.3923/ja.2006.32.36

Abstract

This study shows how a multiple linear regression model can be used to model palm oil yield. The methods are illustrated by examining the time series data of foliar nutrient compositions as one of the independent variable and fresh fruit bunch as dependent variable. Other independent variables include the nutrient balance ratio and major nutrient composition. This modeling approach is capable of identifying the significant contribution of each independent variable in the improving the modeling performance. We find that the quantile-quantile plot demonstrates the existing of outlier and this directs us to use robust M-regression for removing the negative impact of outliers. Results show that robust regression in this case gives a better results than conventional regression in modeling oil palm yield.

Item Type:Conference or Workshop Item (Paper)
Uncontrolled Keywords:oil palm, multiple linear regression
Subjects:Q Science
Divisions:Science
ID Code:21535
Deposited By: Liza Porijo
Deposited On:27 May 2014 03:23
Last Modified:27 May 2014 03:23

Repository Staff Only: item control page