论文总字数：13591字

摘 要

为检测我国农产品期货市场是否具有价格发现能力，本文首先选用相关系数进行大致的观测，进而使用协整检验分析变量关系，接着通过Granger因果模型判断引导性，最后用误差修正模型对数据进行解释和分析。

本文选取了2015-2018这个时间段内，郑州的小麦期货和大连的大豆期货每月中旬的期货结算价格作为样本。现货也选取了对应的时间，大豆和小麦的现货价格来源于中国粮油价格检测系统，分别选取他们每个月的平均价格，为了减少地理位置不同产生的差异，分别选取了对应地区的现货数据。小麦选取了河南地区的数据，大豆选取了黑龙江地区的数据（由于黑龙江与辽宁的地理因素等差异不大且辽宁地区的数据缺失）。通过研究，每月中旬的大豆、小麦的期货结算价格与月平均现货价格之间均存在相关性,期货结算价格能大致推断出现货的月平均价格。且实验结果表明，大豆和小麦的期货都能对现货价格进行预测，即小麦和大豆期货都有价格发现功能。经研究结果表明，小麦期货价格发现功能较大豆期货的价格发现功能较为显著。

关键词：期货价格，现货价格，误差修正模型，Granger因果检验

1

Abstract

In order to test whether China's agricultural futures market has the ability to find prices, this paper first selects the correlation coefficient for general observation, and then uses co-integration test to analyze the variable relationship, then judges the guiding effect through Granger causal model, and finally uses error correction model to interpret the data and analysis.

This paper selects the data of Dalian and Zhengzhou Commodity Exchange. Because only they trade agricultural futures. During the time period of 2015-2018, the futures settlement price of Zhengzhou's wheat futures and Dalian soybean futures was selected as a sample. The spot price was also selected. The spot price of soybean and wheat was derived from the Chinese grain and oil price detection system, and their average price per month was selected. In order to reduce the difference in geographical location, wheat was selected from the Henan region. At the same time, soybeans selected data from the Heilongjiang region, due to the lack of data in Liaoning and the geographical factors of Heilongjiang. Through research, there is a correlation between the futures settlement price of soybeans and wheat in the middle of the month and the monthly average spot price. The futures settlement price can roughly infer the monthly average price of the goods. Finally, through experiments we prove that both have price discovery function.

Key words: Futures price，Present price，Error correction model，Granger causality test

2

目录

摘 要.........................................................................................................1

Abstract...................................................................................................2

第一章 引言......................................................................................4

1.1研究背景………………………………………………………………..4

1.2文献综述………………………………………………………………..4

1.3本文的研究方法，意义………..……………………………………...5

第二章变量、数据选择和研究方法................................................7

2.1样本选取………………………………………………………………..7

2.2研究方法………………………………………………………………..7

2.2.1 相关系数………………………………………………………..7

2.2.2 协整检验………………………………………………………..7

2.2.3 Granger因果检验……………………………………….……..8

2.2.4 误差修正模型…………………………………………………..8

第三章 实证研究............................................................................10

3.1 相关性分析………………………………………………………..10

3.2 协整检验…………………………………………………………..12

3.3 Granger因果检验……………………………………….………..13

3.4 误差修正模型……………………………………………………..14

第四章 结论....................................................................................15

参考文献..........................................................................................16

致 谢................................................................................................17

请支付后下载全文，论文总字数：13591字