用GARCH模型预测股票指数波动率

《用GARCH模型预测股票指数波动率》由会员上传分享，免费在线阅读，更多相关内容在工程资料-天天文库。

1、实用文档用GARCH模型预测股票指数波动率目录Abstract21.引言32.数据63.方法73.1．模型的条件平均73.2.模型的条件方差83.3预测方法93.4业绩预测评价94.实证结果和讨论125.结论16References18标准文案实用文档AbstractThispaperisdesignedtomakeacomparisonbetweenthedailyconditionalvariancethroughsevenGRACHmodels.Throughthiscomparison,totestwhet

2、heradvancedGARCHmodelsareoutperformingthestandardGARCHmodelsinpredictingthevarianceofstockindex.Thedatabaseofthispaperisthestatisticsof21stockindicesaroundtheworldfrom1Januaryto30November2013.Byforecastingone–step-aheadconditionalvariancewithindifferentmodels,

3、thencomparetheresultswithinmultiplestatisticaltests.Throughoutthetests,itisfoundthatthestandardGARCHmodeloutperformsthemoreadvancedGARCHmodels,andrecommendsthebestone-step-aheadmethodtoforecastofthedailyconditionalvariance.Theresultsaretostrengthentheperforman

4、ceevaluationcriteriachoices;differentiatethemarketconditionandthedata-snoopingbias.Thisstudyimpactthedata-snoopingproblembyusinganextensivecross-sectionaldataestablishandtheadvancedpredictiveabilitytest.Furthermore,itincludesa13years’periodsampleset,whichisrel

5、ativelylongfortheunpredictabilityforecastingstudies.ItispartoftheearliestattemptstoinspecttheimpactofthemarketconditionontheforecastingperformanceofGARCHmodels.ThisstudyallowsforagreatchoiceofparameterizationintheGARCHmodels,anditusesabroadrangeofperformanceev

6、aluationcriteria,includingstatisticallossfunctionandtheMince-Zarnowitzregressions.Thus,theresultsaremorerobustanddiffuselyapplicableascomparedtotheearlieststudies.KEYWORDS:GARCHmodels;volatility,conditionalvariance,forecast,stockindices.标准文案实用文档1.引言波动性预测可以运用到投

7、资组合选择，期权定价，风险管理和以波动性为基础的交易策略。GARCH模型族被广泛的运用在模拟预测金融资产的波动性。另一个普遍运用的模式为简单的时间序列模型，例如指数加权移动平均（EWMA）模型和复杂随机波动性模型(PoonandGranger,2003)。对不同金融市场波动性的预测，Ederington在2005年发现GARCH模型通常的表现优异于EWMA模型。同样的，关于随机过程的波动率建模，有强有力的证据证明随机波动模型的样品性能堪比GARCH模型（FlemingandKirby,2003）.标准GARCH模型

8、于1986年被Bollerslev提出后，为了规范条件方差，更多复杂的GRACH模型参数被提出。这些先进的GARCH模型试图去更好的捕捉经验主义观察到条件方差的过程。例如，EGARC模型，GJR模型，TGARCH模型和NGARCH模型获得的负返回流的非对称性效应。更为广义的参数化，像APARCH模型和HGARCH模型，包含大量较为简单的GARCH模型（Zak