文本描述
摘要 中国期货市场经历了近30年的快速发展期,各种交易品种日益丰富,交易投 资活动日趋活跃。我国商品期货市场的高速发展进一步促进了我国商品现货市场 的繁荣稳定,同时也为我国各期货品种产业链上下游的关联公司和机构、各界市 场参与者提供了广泛的风险防范和套期保值途径。大豆、豆粕和豆油三者每一个 品种在我国国民经济中的都占有着重要地位,以三种农产品中每一个为圆心都聚 集和围绕了众多的上下游企业并构成产业链,因此这三者每一个价格的大幅变化 都对这些企业的生产产生了重大影响。而期货市场的诞生就为这些相关产业链企 业提供了依靠期货规避价格波动风险和套期保值的途径,期货价格的大幅变化也 促使各界研究者对期货市场的波动性和波动性溢出效应进行更深的探索。鉴于此, 本文试图通过实证研究大豆,豆粕和豆油三种期货品种的波动溢出效应,对产业 链中的公司和投资者使用期货套期保值和跨品种套利对冲,并为决策者和监管机 构制定宏观经济政策提供参考。 本文选取了大豆、豆粕和豆油期货主力合约指数从2006年1月9日到2018 年9月21日12年半共3094个交易日的收盘价数据,取对数收益率并进行100倍 扩大后,再对三者的价格对数收益率序列进行描述性统计与平稳性检验。在确定 序列最优滞后阶数后构建VAR模型,并进行格兰杰因果检验基于均值层面研究序 列间的价格波动传导,最后构建三元BEKK-GARCH(1,1)模型,基于方差层面 求证大豆、豆粕和豆油期货三者的价格对数收益率序列之间的波动溢出关系。 研究结果表明:大豆、豆粕和豆油期货的价格对数收益率序列均存在尖峰性 和厚尾性,三个序列都接近但不完全服从正态分布。三个序列均存在显著的波动 聚集性。三个序列均存在着显著的ARCH效应,用GARCH族模型能更好地研究 分析三者的波动溢出效应。大豆、豆粕和豆油期货三者的价格对数收益率序列两 两间均存在单向的波动溢出效应,并且也存在双向的波动溢出效应。三种农产品 期货市场之间的波动溢出效应相互影响的强弱大小也不同,豆油期货的价格变化 对于大豆期货和豆粕期货有着更为强劲的影响,大豆期货的价格变化对于豆粕期 货也有着更为强劲的影响,影响的程度大小可以概括为:豆油大于大豆和豆粕, 大豆大于豆粕。根据这一得出的实证结论,最后本文给期货投资者和农业经济政 策制定者提出了一些建议和启发以供参考。 关键词:农产品期货,BEKK-GARCH模型,波动溢出效应 iii 北京交通大学硕士专业学位论文ABSTRACT ABSTRACT China'sfuturesmarkethasexperiencedarapiddevelopmentperiodofnearly30 years,withincreasinglyrichtradingvarietiesandincreasinglyactivetradingand investmentactivities.TherapiddevelopmentofChina'scommodityfuturesmarket furtherpromotestheprosperityandstabilityofChina'scommodityspotmarket,andat thesametimeprovidesawiderangeofriskpreventionandhedgingapproachesforthe relatedcompaniesandinstitutionsintheupstreamanddownstreamofChina'sindustrial chainoffuturesvarietiesandmarketparticipantsfromallwalksoflife.Eachvarietyof soybean,soybeanmealandsoybeanoiloccupiesanimportantpositioninChina's nationaleconomy.Eachofthethreeagriculturalproductsgathersandsurrounds numerousupstreamanddownstreamenterprisesandconstitutesanindustrialchainat thecenterofthecircle.Therefore,thesignificantchangesinpricesofeachofthese threeproductshaveasignificantimpactontheproductionoftheseenterprises.Thebirth ofthefuturesmarketprovidestheseindustrialchainenterpriseswithawaytoavoidthe riskofpricefluctuationandhedgebyfutures.Inviewofthis,thispapertriestostudy thevolatilityspillovereffectofthreefuturesvarietiesofsoybean,soybeanmealand soybeanoilthroughempiricalresearch,toprovidereferenceforcompaniesand investorsintheindustrialchaintousefutureshedgingandcross-varietyarbitrage hedging,andtoprovidereferenceforpolicymakersandregulatorstoformulate macroeconomicpolicies. Thispaperselectedtheclosingpricedataof3094tradingdaysofthemaincontract indexofsoybean,soybeanmealandsoybeanoilfuturesfromJanuary9,2006to September21,2018for12andahalfyears.Aftertakingthelogarithmicreturnrateand extendingitby100times,descriptivestatisticsandstationarytestwereconductedon thelogarithmicreturnrateseriesofpricesofthethree.Afterdeterminingtheoptimallag orderofthesequence,theVARmodelwasbuilt,andthegrangercausalitytestwas carriedouttostudythepricefluctuationtransmissionamongthesequencesbasedonthe meanlevel.Finally,aternarybekk-garch(1,1)modelwasbuilt,andthevolatility spilloverrelationshipamongthepricelogyieldseriesofsoybean,soybeanmealand soybeanoilfutureswasverifiedbasedonthevariancelevel. Theresultsshowthatthelogarithmpriceyieldseriesofsoybean,soybeanmeal andsoybeanoilfuturesallhavesharppeaksandthicktails.Allthreesequenceshave iv 北京交通大学硕士专业学位论文ABSTRACT significantwaveaggregation.ThethreesequenceshavesignificantARCHeffect,and thefluctuationspillovereffectofthethreesequencescanbebetterstudiedandanalyzed byusingGARCHfamilymodel.Thereisone-wayvolatilityspillovereffectintheprice logyieldseriesofsoybean,soybeanmealandsoybeanoilfutures,andthereisalso two-wayvolatilityspillovereffect.Threekindsofagriculturalproductfuturesmarket volatilityspillovereffectbetweenthestrengthoftheinteractionofdifferentsizesand soybeanfuturespricechangesforsoybeansandsoybeanmealfutureshasastronger influence,soybeanfuturespricechangesforsoybeanmealfuturesalsohasastronger influence,theextentoftheimpactofthesizecanbesummarizedas:soybeanoilismore thansoybeanandsoybeanmeal,soybeanisgreaterthanthemeal.Accordingtothe empiricalconclusionofthispa