干散货航运即期与远期运价的联动关系及溢出效应研究

发布时间：2018-03-23 12:48

本文选题：VAR模型　切入点：非对称BEKK模型　出处：《中国海洋大学》2014年硕士论文

【摘要】：随着我国社会建设步伐的加快，我国所需要的铁矿石等大宗货物的进口量逐年增加，使得我国铁矿石的航运量加大，与之相伴的就是在整个航运过程中所要面临的风险在逐渐扩大。如何运用金融衍生品自身的特点进行风险的规避一直是船东和货主亟待解决的问题，尤其是干散货这种运价变化多样的商品更需要进行深入的研究，在这种情况下，远期运费协议FFA应运而生。在进行即期市场操作时，船东和货主可以使用FFA做一个反向的对冲交易，以规避运价波动带来的风险。本文以C3、C5即远期航线的收益率数据为研究对象，采用向量自回归模型和非对称BEKK模型对FFA的信息传递效率和价格预测方式进行研究。其中C3、C5航线属于波罗的海海岬型船型BCI，C3航线运输路径是从图巴朗港到北仑或宝山，运载量是16万公吨，C5航线运输路径是从澳西到北仑或宝山，运载量是15万公吨，两条航线主要运输铁矿石，报价形式都采用美元/吨。本文主要从以下两个方面进行论述：首先，，以向量自回归模型为基础，利用脉冲响应分析和方差分解分析，对C3、C5航线的即期与远期市场及其航线间的联动关系进行研究。脉冲响应结果显示：同一航线即期市场的波动会引起远期市场的波动，反之则影响较小；C5航线的波动能迅速传给C3航线，反之则较弱。通过方差分解分析可以发现，即期市场的预测波动大约有50%来源于对应的远期市场，而远期市场的预测波动主要取决于远期市场本身；C3航线的预测波动主要源于本航线而C5航线的预测波动则有一半来源于C3航线。其次，建立非对称BEKK模型，运用ARCH检验和Wald检验对C3、C5航线的波动溢出结果进行分析。ARCH检验得出C3、C5航线的即远期收益率存在着ARCH效应；BEKK的估计结果得出C3、C5航线间的波动溢出效应确实会受到经济环境变化的影响，金融危机前期，C3航线即期市场信息流通速率明显弱于其余其他市场，C5航线间的波动溢出效应更加明显，金融危机中期，由于C5航线的信息量比C3航线充足，因此信息传播并未受到影响，金融危机后期，航线间的交叉作用逐渐取代自身滞后项影响，航线间信息流通速率加快。Wald检验得出，在金融危机前期C3即期与C5远期之间不存在波动溢出效应；在金融危机中期C3航线的即远期之间不存在波动溢出效应，C3与C5航线的远期之间不存在波动溢出效应；金融危机后期C3、C5航线之间均存在波动溢出效应。综上分析，在对FFA的信息传递速率进行研究的时候，首先考虑航线自身的历史信息以及相似航线即远期波动的影响，同时分析数据所处的经济时期；对FFA的价格预测方式进行研究的时候，要统计所有类似航线的数据信息加以研究。
[Abstract]:With the acceleration of the pace of social construction in China, the import of iron ore and other bulk goods needed by our country has increased year by year, which has led to an increase in the volume of shipping iron ore in China. At the same time, the risks to be faced in the whole shipping process are gradually expanding. How to use the characteristics of financial derivatives to avoid risks has always been an urgent problem for shipowners and shippers. In particular, dry bulk goods, which have varied freight rates, need to be studied deeply. In this case, the forward freight agreement (FFA) emerges as the times require. Shipowners and consignors can use FFA to make a reverse hedging transaction to avoid the risk caused by the fluctuation of freight rates. Vector autoregressive model and asymmetric BEKK model are used to study the information transmission efficiency and price prediction mode of FFA. Among them, route C3C5 belongs to the Baltic promontory type BCIC3 route, which is from Port Tubaron to Beilun or Baoshan. The transportation route is from Australia to Beilun or Baoshan with a capacity of 150000 metric tons. Iron ore is mainly transported on both routes in the form of US dollars / ton. This paper mainly discusses the following two aspects: firstly, based on the vector autoregressive model, impulse response analysis and variance decomposition analysis are used. The relationship between spot market and forward market of C3C5 route and the linkage between them are studied. The results of pulse response show that the fluctuation of spot market on the same route will cause volatility of forward market. On the contrary, the fluctuation of C5 route with smaller influence can be quickly transmitted to the C _ 3 route, otherwise it is weaker. Through variance decomposition analysis, it can be found that about 50% of the forecast volatility of spot market comes from the corresponding forward market. The forecast fluctuation of forward market mainly depends on the forecast fluctuation of the C _ 3 route of the forward market itself, and half of the forecast fluctuation of C _ 5 route comes from the C _ 3 route. Secondly, an asymmetric BEKK model is established. By using ARCH test and Wald test to analyze the volatility spillover result of C3C5 route. Arch test shows that there is ARCH effect on the forward return rate of C3C5 route. The result shows that the volatility spillover effect between C3C5 routes will really be affected. The impact of environmental change, In the early stage of the financial crisis, the current information flow rate of C _ 3 route was obviously weaker than that of the other markets. In the middle of the financial crisis, the amount of information on the C _ 5 route was more abundant than that on the C _ 3 route. Therefore, the information dissemination has not been affected. In the later period of the financial crisis, the cross-action between routes gradually replaced the influence of their own lag term, and the speed of information flow between routes was accelerated. Wald test shows that, There is no volatility spillover effect between C _ 3 spot and C _ 5 forward in the early stage of financial crisis, and there is no volatility spillover effect between C _ 3 and C _ 5 routes in the middle of financial crisis, and there is no volatility spillover effect between C _ 3 and C _ 5 route in the middle of financial crisis. In the latter stage of the financial crisis, there are volatility spillover effects between C _ 3 C _ 5 and C _ 3 C _ 5 routes. On the basis of the above analysis, when we study the information transfer rate of FFA, we first consider the historical information of the route itself and the influence of the similar routes, that is, the long-term fluctuations, and analyze the economic period in which the data are located. When we study the price forecasting method of FFA, we should count all the data information of similar routes.
【学位授予单位】：中国海洋大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：F552;F224;F426.1

【参考文献】