我國豆粕基差波動的影響因素與交易策略研究

Factors and Trading Strategies of Soya Meal Futures in China

Student thesis: Doctoral Thesis

View graph of relations

Author(s)

Related Research Unit(s)

Detail(s)

Awarding Institution
Supervisors/Advisors
  • Tao LI (Supervisor)
  • Yifan Xu (External person) (External Supervisor)
Award date24 Oct 2023

Abstract

豆粕是大豆壓榨後的主要產品之一,主要用於飼料和食品工業領域。近年來,伴隨我國豆粕生產和消費的快速增長,行業內企業普遍採用期貨工具進行套期保值。然而,由於時間、區域市場等差異性,豆粕現貨價格和期貨價格在大部分時間並不完全匹配,簡單地進行套期保值並不能完全規避價格風險。

現貨價格與對應期貨價格的差值為基差,通過設計場外基差合約並進行交易,可以規避期現貨價格不匹配帶來的風險。基差交易具有保證金佔用少、波動幅度小、應用靈活、能有效連接產業上下游等優點,目前已經成為我國豆粕市場主流交易形式。研究我國豆粕基差波動的影響因素並進行應用,不僅對相關企業的經營決策具有實踐指導意義,而且對國內糧油行業和飼料行業的健康發展具有重要意義。

豆粕基差的影響因素眾多,可以從不同角度進行分析。首先,根據經典基差倉儲理論,影響豆粕基差的因素是各類儲運成本,對應到我國大豆進口為主的實際,大豆進口相關成本將影響到豆粕基差;其次,豆粕現貨市場的供需關係同時影響現貨價格與期貨價格,但對二者的衝擊不同,因而構成豆粕基差的基礎和直接的影響因素;最後,市場的交易情況也影響了基差的波動。通過不同角度研究多種因素對我國豆粕基差波動的影響以及互相作用機制,掌握其變化規律,將有助於企業經營和行業發展。

本文綜合利用企業實際經營資料和市場公開資料,採用時間序列分析方法對我國豆粕基差波動的相關影響因素進行了研究。

首先,研究了我國豆粕基差的時間序列特徵及應用。在驗證我國豆粕基差為平穩時間序列的基礎上,構建了ARMA(2,1)模型進行基差預測。同時,對我國豆粕現貨市場和期貨市場間的關係進行了分析,研究表明,儘管我國豆粕期貨市場具有價格發現功能,即豆粕現貨價格和期貨價格存在長期協整關係,但二者不存在相互的價格波動溢出關係,豆粕期貨價格對現貨價格存在波動溢出影響,反之則不存在。

其次,從不同角度研究各因素對我國豆粕基差的影響。第一,基差倉儲理論的擴展與驗證。針對我國大豆進口為主的特點,驗證了相關成本因素,包括海運運費、利率、匯率等,與我國豆粕基差具有顯著相關性。第二,基差與供需基本面的關係。在研究基差與成本差、利潤差、盤面榨利和現貨榨利間的Granger因果關係的同時,重點研究了豆粕現貨庫消比、豆加粕庫消比和庫存量等代表供需基本面的變數與基差間的關係。研究表明,豆粕現貨庫消比在小於4的情況下,與豆粕基差互相構成Granger因果關係,豆加粕庫消比同樣於與豆粕基差互相構成Granger因果關係,而豆粕庫存量僅與豆粕基差構成單向的Granger因果關係。第三,基差與市場交易情況的關係。研究表明,換手率與基差間為單向Granger因果關係,但增倉比與基差間無顯著關係。

最後,研究豆粕基差與相關商品基差的關係。主要分析了豆粕基差與進口大豆基差和豆油基差間的關係。研究表明,進口大豆CNF升貼水構成對豆粕基差的Granger因果關係,反之則不然。另外,豆粕基差與豆油基差雖然互不構成Granger因果關係,但具有顯著負相關性。

基於上述對我國豆粕基差的研究,本文對相關企業的基差交易策略進行了優化,包括基於時間序列預測模型的交易優化、基於影響因素分析的交易優化和跨商品基差套利策略等,以實現基差交易的改進。
Soybean meal is one of the main products after soybean pressing, mainly used in the feed and food industry. In recent years, with the rapid growth of soybean meal production and consumption in China, enterprises in the industry have generally used futures tools for hedging. However, due to differences in time and regional markets, the spot prices and futures prices of goods do not fully match for most of the time, and simply hedging cannot completely avoid price risk.

The difference between the spot price and the corresponding futures price is the basis, which can be avoided by designing corresponding over-the-counter contracts for trading. Basis trading has the advantages of low margin occupation, low volatility and frequency, flexible application, and effective connection between upstream and downstream industries. It has become the mainstream trading form in the soybean meal market. Studying the influencing factors of the fluctuation of soybean meal basis in China and applying them not only has practical guiding significance for the business decision-making of relevant enterprises, but also has important significance for the healthy development of the domestic grain and oil industry and feed industry.

There are many influencing factors on the basis difference of soybean meal, which can be analyzed from different perspectives. Firstly, according to the classical basis warehousing theory, the factors that affect the basis of soybean meal are various storage and transportation costs. Correspondingly, in China, where soybean imports are the main source, the costs related to soybean imports will affect the basis of soybean meal; Secondly, the supply and demand relationship in the soybean meal spot market affects both spot and futures prices, but the impact on both is different, thus forming the basis and direct influencing factor of the soybean meal basis; Finally, the trading situation in the market also affects the volatility of the basis. Studying the impact and interaction mechanisms of various factors on the fluctuation of soybean meal basis in China from different perspectives and mastering their changing patterns will be helpful for enterprise management and industry management.

This paper comprehensively utilizes actual business data of enterprises and publicly available market data, and uses time series analysis method to study the relevant factors of domestic soybean meal basis fluctuation.

Firstly, study the time series characteristics and applications of soybean meal basis differences in China. On the basis of verifying that the basis of soybean meal in China is a stationary time series, an ARMA (2,1) model was constructed to predict the basis. At the same time, an analysis was conducted on the relationship between China's soybean meal spot market and futures market. The study showed that although China's soybean meal futures market has a price discovery function, that is, there is a long-term cointegration relationship between soybean meal spot prices and futures prices, but there is no symmetrical price volatility spillover relationship between the two. Soybean meal futures prices have a volatility spillover effect on spot prices, and vice versa.

Secondly, study the impact of various factors on the basis difference of soybean meal in China from different perspectives. Firstly, the extension and validation of the basis warehousing theory. Based on the characteristics of soybean imports in China, relevant cost factors, including sea freight, interest rates, exchange rates, etc., are verified to have a significant correlation with the basis difference of soybean meal in China. Secondly, the relationship between basis and supply and demand fundamentals. While studying the Granger causal relationship between basis and cost difference, profit difference, disk profit and spot profit, the focus was on studying the relationship between variables representing supply and demand fundamentals, such as soybean meal spot inventory consumption ratio, soybean meal inventory consumption ratio, and inventory quantity, and basis. Research has shown that when the spot stock to consumption ratio of soybean meal is less than 4, it forms a Granger causality with the soybean meal basis, while the soybean meal stock to consumption ratio and inventory can also represent the supply and demand fundamentals, forming a Granger causality with the soybean meal basis. Thirdly, the relationship between basis and market trading conditions. Research has shown that there is a one-way Granger relationship between turnover rate and basis, but there is no significant relationship between increase in position ratio and basis.

Finally, study the relationship between soybean meal basis and related commodity basis. Mainly studied the relationship between soybean meal basis and imported soybean basis and soybean oil basis. Research has shown that the CNF premium on imported soybeans constitutes a Granger causality relationship with the basis, while the basis does not constitute a Granger causality relationship with the CNF premium on imported soybeans. In addition, although the soybean meal basis and soybean oil basis do not form a Granger causality relationship, they have a significant negative correlation.

Based on the above research on the basis of soybean meal in China, this article has optimized the basis trading strategies of relevant enterprises, including trading optimization based on basis prediction models, trading optimization based on supply and demand fundamentals, and cross commodity basis arbitrage strategies, all of which can achieve improvement in basis trading.

    Research areas

  • soybean meal basis, Time series, Supply and demand fundamentals, Trading strategy