联系我们

论文查询结果	返回搜索

论文编号：	11892
作者编号：	2120182997
上传时间：	2020/6/23 19:48:33
中文题目：	期货市场套期保值比率研究 ——基于分位数回归法
英文题目：	Research of hedge ratio in futures market: Based on quantile regression method
指导老师：	王永进
中文关键字：	套期保值比率；分位数回归法；价格风险
英文关键字：	Hedge ratio; Quantile regression; Price risk
中文摘要：	随着经济全球化程度的不断加深，我国与世界各国的联系越来越紧密，我国商品市场的价格受世界范围内各种因素的影响也在不断加深。与此同时，我国期货市场的成交金额和成交量也在近些年有显著的提升，期货市场日趋成熟。在商品价格波动日趋剧烈的今天，商品持有者通过期货市场来对冲现货市场价格风险的需求不断加大，因此套期保值比率的选择问题也日趋重要。 本文首先梳理了与套期保值比率相关的国内外文献，之后介绍了几种套期保值模型，包括传统OLS模型、VAR模型、ECM模型，随后引入了分位数回归法并将其运用到套期保值比率的分析中去。本文选取我国期货市场中上市时间较早、交易规模较大、市场定价能力较强的八个交易品种，利用近十年的数据进行研究分析，使用不同的模型来估计套期保值比率；除了利用日数据估计套期保值比率外，本文还使用单周、双周、月数据来计算不同期限的套期保值比率，并分析期限对套期保值比率的大小以及套期保值效果的影响；最后，使用风险最小化的方法对现货收益率位于不同分位数下的套期保值效果进行评价。 本文着重分析了分位数回归法的估计结果。最终结果显示，使用OLS模型估计出的套期保值系数小于使用VAR模型估计出的套期保值系数，VAR模型估计出来的系数又小于ECM模型估计出来的系数；随着套期保值期限的不断增长，使用各种模型估计出来的系数都有所扩大；使用分位数回归法估计出的股指期货套期保值比率呈现明显的倒U型分布，商品期货中天然橡胶、黄金、镍期货也呈现倒U型分布，而更长期限的套期保值比率则没有非常明显的规律；在对模型套期保值效果的检验中我们发现，利用期货市场对现货市场进行套期保值能显著降低投资者面临的风险，且大部分合约的分位数套期保值比率在现货收益率处于极端行情中的表现都优于其他几个模型估计出的套期保值比率的表现，这对投资者在极端情况下进行套期保值比率的选择有一定的指导意义。
英文摘要：	With the development of economic globalization, China is more and more closely connected with other countries in the world. At the same time, the transaction amount and trading volume of China's futures market have also increased significantly in recent years, and the futures market is becoming more and more mature. Nowadays, with the commodity price fluctuation becoming more and more violent, the demand of commodity holders to hedge the price risk of spot market through futures market is increasing, so the choice of hedge ratio is becoming more and more important. In this paper, we first sort out relevant domestic and foreign literatures related to hedge ratio, and then introduce several hedge models, including traditional OLS model, VAR model and ECM model. Then we introduce quantile regression method and apply it to the analysis of hedge ratio. In this paper, we select eight future contracts in China's futures market which have longer listing time, larger trading scale and stronger market pricing ability, and then use nearly ten years’ data to do the research. Then we use different models to estimate the hedge ratio. Besides using the daily data to estimate the hedge ratio, this paper also uses the data of one week, two weeks and one month to calculate the hedge ratio of different periods, and analyzes the influence of the period on the size of the hedge ratio and the hedge effect. Finally, the hedge effect of spot yield under different quantile is evaluated by means of risk minimization. This paper focuses on the estimation results of quantile regression method. The final results show that the hedge coefficient estimated by OLS model is smaller than that estimated by VAR model, and the coefficient estimated by VAR model is smaller than that estimated by ECM model. With the increasing of hedge period, the coefficients estimated by various models are all enlarged. The hedge ratio of stock index futures estimated by quantile regression shows an obvious inverted u-shaped distribution, and the natural rubber, gold and nickel futures also show an inverted u-shaped distribution, while the hedge ratios of longer term have no obvious rule. In the test of hedge effect, we found that using the futures market to hedge the risks of the spot market can significantly reduce the risk of investors, and in extreme case, most contracts’ quantile hedge ratios have the best performance. This can give some advice for the investors in extreme circumstances to choose the best hedge ratio.
查看全文：	预览  下载(下载需要进行登录)