Conjugate Gradient (CG) methods have an important role in solving large
scale unconstrained optimization problems. Nowadays, the Three-Term CG method has
become a research trend of the CG methods. However, the existing Three-Term CG
methods could only be used with the inexact line search. When the exact line search
is applied, this Three-Term CG method will be reduced to the standard CG method.
Hence in this paper, a new Three-Term CG method that could be used with the exact
line search is proposed. This new Three-Term CG method satisfies the descent condition
using the exact line search. Performance profile based on numerical results show that
this proposed method outperforms the well-known classical CG method and some related
hybrid methods. In addition, the proposed method is also robust in term of number of
iterations and CPU time.
Regression is one of the basic relationship models in statistics. This paper focuses on the formation of regression models for the rice production in Malaysia by analysing the effects of paddy population, planted area, human population and domestic consumption. In this study, the data were collected from the year 1980 until 2014 from the website of the Department of Statistics Malaysia and Index Mundi. It is well known that the regression model can be solved using the least square method. Since least square problem is an unconstrained optimisation, the Conjugate Gradient (CG) was chosen to generate a solution for regression model and hence to obtain the coefficient value of independent variables. Results show that the CG methods could produce a good regression equation with acceptable Root Mean-Square Error (RMSE) value.