Identifying Factors Affecting Palm Oil Prices Based On Grey Incidence Analysis

Palm oil industry is an industry that supports Malaysian economy. It triggers economic activities, but due to the uncertainty in the world of economic force, the palm oil prices in Malaysia have a high volatility movement. The volatility movement of palm oil prices is due to many factors such as the prices of other vegetables oil and crude oil. The purpose of this research is to find the main factors that affecting crude palm oil price quantitatively and to overcome the problem by providing an early warning signal in order to prevent huge losses in the future. By employing absolute grey incidence analysis and also correlation, this study calculates the grey incidence degree to determine the cointegration of palm oil prices with other factors which are soybean oil, rapeseed oil, palm kernel oil, coconut oil and crude oil prices towards palm oil prices. It demonstrates that palm kernel oil price has a high relationship between the palm oil prices.


Introduction
Oil palm plays an important role in supporting Malaysia's economy. On top of that, Malaysia has gained position as the world leading palm oil producer (Goniur,2011). The palm oil industry formed the economic backbone of Malaysia with the total export of palm oil in Malaysia until the end of year 2010 was 2,064, 247 Millions/Tonne (M/T). It was increased from the previous year, 2009 which was 1,892,099 M/T. Currently Malaysia accounts for 39% of world palm oil production and 44% of world exports. Edward, Hondgon, Hamer, Mitchell, Ahmad, Cornel and Wilcove (2010) state that there are countries those believe and prefer to use palm oil amongst the vegetable oil because it is healthier. As a result, they increase consumption of palm oil in food products. Palm oil is the most popular agricultural commodity in many countries. This industry makes significant contributions in the Malaysian economic growth and development. Palm oil also is commonly used in food industry in Malaysia. According to Shakila, Noryati and Maheran (2017) food industry has low stock volatility that benefits the Malaysia's economy. As emphasized by Nordin and Ismail (2014) the stock market index in Malaysia is influenced by the prices of palm oil. Other than that, the widely used of palm oil made researchers interested to explore more on oil palm (Syamsiah et al., 2011;Roseleena et al., 2011).
Other considered factors in this study are soybean oil prices, rapeseed oil prices, palm kernel oil prices, coconut oil prices and Brent crude oil prices. In this research, Grey incidence analysis is selected as a method to evaluate the correlation or interaction degree with other variables in the system. Deng Ju-long is the first person that proposed the concepts and establishes the grey system in 1982. More specifically, in 1982, Deng Ju-Long published the first research paper in the area of grey systems in the international journal entitle Systems and Control letters, published by North-Holland Co. His paper was titled "Control problems of Grey Systems".
The publication of this paper signaled the birth of grey systems theory after many years of effective research of the founding father (Liu, Fang,& Lin, 2006). According to Deng the output of the system is more important compared to the input and in addition both grey quantity and grey relationship within the systems are also important (Ping & Keija,2005).The fundamental idea of grey incidence analysis is that the closeness of a relationship is judged based on the similarity level of the geometric patterns of sequences curves . There are many incidence models such as Deng's degree of grey incidence, absolute degree of grey incidence, relative degree of grey incidence, grey incidence of B-model, grey incidence of entropy and many more models afterwards (Liu, Forrest & Yang, 2012).

Literature Review Palm Oil Prices
There are many studies proposed in analyzing the factors affecting palm oil prices. Different study conducted to attest the factors contribute to the movement of palm oil prices. In examining the correlation of crude palm oil and other factors, Chuangchid, Wiboonpongse, Sriboonchitta and Chaiboonsri (2012) discovered that there is co-integration relationship between palm oil prices with soybean oil prices and crude oil prices. The soybean oil prices are found to have some dependence with the growth rate of palm oil, while crude oil has a fairy weak dependence. They use the bivariate extreme value in investigating the relationship between prices of palm oil and crude oil.
In a different study, Sanders, Balagtas and Gruere (2012) conduct a study to investigate the factors leading to the growth in palm oil prices. The selected factors are soybean oil prices and crude oil prices. They discover that, there is a relationship between both factors and palm oil prices. However, the interaction of palm oil and crude oil prices are negative. In addition, their study also shows that a soybean oil market is a vital driver of the palm oil boom but not for crude oil.

Grey System
Liu, Forrest and Yang (2012) discover that probability and statistics, fuzzy mathematics and also grey systems theory are among the methods and theories used in studies of non-deterministic systems. Even though they study the objects with difference uncertainties, the commonality of all these theories is to make meaningful sense out of incompleteness and uncertainties. The fundamental meaning of grey system is by having partially complete information. The comparisons between grey system theory, probability statistics, rough sets and fuzzy mathematics are tabulated below. Mathematics. Grey Incidence Analysis Grey incidence analysis is put forward by Professor Deng Julong in grey system theory. The grey incidence analysis is an important part in grey systems theory. According to Li-zhi and Si-feng (2009), grey incidence analysis is used to know the mutual reaction among the factors, since it can determine the development situation and tendency of the systems. It is a kind of system analysis method which applies the degree of grey incidence to weight the relationship between system characteristics factor and system behavior factor quantitatively (Wang & Dang, 2009). The fundamental idea of grey incidence analysis is that the closeness of a relation is judged based on the similarity patterns of sequences curves. It indicates, their relationship is closer. The more similar the curves are, the higher the grey incidence degree between the sequences and vice versa. There are many types of grey system such that Grey Incidence Analysis, Grey Cluster and Grey Statistical Evaluations, Grey Systems Modeling, Grey Combined Models, Grey Prediction and many more (Liu & Yin, 2006). The calculation procedure of the absolute degree for grey incidences is as follows:

Grey systems theory
4) Compute the absolute degree of grey incidences (8) The Application of Grey Incidence Degree Suppose palm oil prices as a system's characteristic sequences and all other variables as systems of behavioral sequence. First is to constitute the system characteristic sequences and behavioral sequence by applying equations (1) and (2). The value in equation (2) represents the factors that affect palm oil prices such as soybean oil, rapeseed oil, palm kernel oil, coconut oil and Brent crude oil. Whereas, the number of considered factors and in this study is 5.   (3) and (4). Here D is called a zero starting point operator with D X i as the image of zero starting point of i X . The variable 0 i X means the notation for zero starting point image for each factor. i refers to the considered factor. Here each data in the series of data minuses the 1 st input data in that series. 0 0 X means the notation for zero starting image for palm oil prices.
The 2 nd input in palm oil price data subtracts the 1 st input data. The 3 rd , 4 th , 5 th and all data in palm oil data series minuses the 1 st input data. Whereas 0 1 X refers to the notation for zero starting image for soybean oil prices and the same goes to rapeseed oil prices, palm kernel oil prices, Brent crude oil prices and coconut oil prices. In this study, i X and 0 X are 1-time interval sequences of the same length. In order to find grey incidence degree we shall find | | | |, 0 i S S and| 0 S S i  | for all factors by using (5), (6) and (7) respectively. Then, the absolute degrees of grey incidence were computed by using equation (8). The degrees of grey incidence are as follows i. 01  = The grey incidence degree for soybean oil prices ii. 02  = The grey incidence degree for rapeseed oil prices iii. 03  = The grey incidence degree for palm kernel oil price.
iv. 04  = The grey incidence degree for coconut oil prices v. 05  = The grey incidence degree for crude oil prices According to Yuan, Liu and Chen (2010), the grey incidence degrees i 0  were measured from . In the first stage, grey incidence degree was evaluated when 0  L . In order to compute grey incidence degree, by using data from 25/10/2012 to 13/11/2013 for both sets of data (Rahman,2012), we took the palm oil prices as a system characteristic sequences or called it as benchmark sequences and marked it as 0 X . Then took the sequences of considered factors, e.g., soybean oil prices as a system behavior sequences and marked it as 1 X .
According to the formula of absolute degree of grey incidence, we could work out the absolute degree of incidence between 0 X and 1 X that expressed it as 01  . The zero starting point was evaluated by using (3) and (4) in order to determine the grey incidence degree. Then, the values of | 1 Therefore it is confirmed that the grey incidence degree is measured from This means the grey incidence degree is never being below than 0.5. From Table 2, under the same time, all the absolute grey incidence degree is from 0.58 and above which is from

Conclusion
From this study, it is found that, whenever i 0  increases or close to 1 then the prices in palm oil and the variables is directly proportional, which means they move in the same direction. This demonstrates the relationship between palm oil and the considered factor prices are highly co-integrated.
The palm kernel oil and palm oil prices were really close, grey incidence degree reaches to 0.8451, followed by coconut oil prices and soybean oil prices with 0.8213 and 0.6001 respectively. Besides that, the correlation of palm kernel oil is also quite high which is 0.6555. This indicates there is a quite strong positive correlation. Correlation for soybean oil and coconut oil is -0.2929 and 0.5352 respectively. Rapeseed oil prices with 0.5874 degree and the correlation is -0.1785 also with negative correlation. This means when the prices of palm oil go up the prices of rapeseed oil goes down. Brent crude oil also shows a negative relationship with palm oil prices with correlation about -0.1795. The work attests that there are many factors involved in the movement of palm oil prices. In this research work, we have empirically verified the factors that affecting palm oil in Malaysia that are soybean oil prices, rapeseed oil prices, palm kernel oil prices, coconut oil prices and crude oil prices towards the palm oil prices. The aim of the study is to ascertain the impact of price volatility of soybean oil prices, rapeseed oil prices, palm kernel oil prices, coconut oil prices towards the palm oil prices. In the study, it was found that coconut oil prices and palm kernel oil prices have a significant positive relationship with the crude palm oil, whereas soybean oil, rapeseed oil and Brent crude oil have a significant negative relationship with crude palm oil prices. The most vital factor that gives huge impacts on palm oil price volatility is palm kernel oil prices followed by coconut oil prices, soybean oil prices, rapeseed oil prices and the last is Brent crude oil prices. The grey incidence value and correlation of each variable are tabulated in Table 1. The grey incidence rank in this study is