臺灣博碩士論文加值系統

本文中，我們引用了一個混合常態模型來分析股票市場跟選擇權市場的關係。觀察在加入隱含波動率的訊息之後，混合常態模型的波動率是否有影響，而我們檢驗的方式是建立信賴區間去看他的變化，在模擬跟實證都是使用的是貝氏估計來探討，最後發現加入隱含波動率這個動作，確實減少了混合常態模型整體
波動率的信賴區間長度，估計值也變精準，也說明了選擇權跟股票市場在混合常態模型估計下會互相影響。

In this paper, we use a mixture normal model to analyze the relationship between the stock market and the option market. Observe after adding the implied volatility, whether the volatility of the mixture normal model has an effect and the way we test is to establish a confidence interval to see its changes. In the simulation and empirical study are using the Bayesian estimate to explore. Finally found to join the implied volatility, does reduce the confidence interval length of total volatility and estimates are more accurate, also shows that the option and the stock market under the mixture normal model will affect each other.

摘要 i
Abstract ii
誌謝 iii
1 Introduction 1
2 MNModelandImpliedVolatility 3
2.1 MixtureNormalModel.............................3
2.2 ImpliedVolatility................................6
3 Theparametersestimation 7
3.1 Bayesianwithstock..............................7
3.2 Bayesianwithstockandoption........................10
4 Simulations 13
4.1 Step.......................................13
4.2 Simulationresults................................17
5 Empiricalstudy 20
5.1 Datadescription................................20
5.2 Empiricalresult.................................21
6 Conclusion 25
References 26

[1] Alexander,C.,Narayanan,S.,(2001)OptionPricingwithNormalMixtureReturns
ModellingExcessKurtosisandUncertaintyinVolatility. ICMA CentreDiscussion
PapersinFinance.
[2] Ehlers,R.S.,(2003)BayesianInferenceandMCMCmethods:nopain,noWindows.
[3] Efron,B.(2013)EmpiricalBayesmodeling,computation,andaccuracy.
[4] Gen¸cay,R.,Salih,A.,(2003)DegreeofMispricingwiththeBlack-ScholesModel
and NonparametricCures. ANNALS OFECONOMICSANDFINANCE, 4,73-101.
[5] Huang,Y.L.(2015)BayesianReliabilityAnalysisofConstant-StressAccelerated
Degradation BasedonGammaProcesswithRandomEffectandTime-ScaleTrans-
formation.
[6] Ismail,M.T.,Phoong,S.Y.,(2015)AComparisonBetweenBayesianandMaximum
LikelihoodEstimationsinEstimatingFiniteMixtureModelforFinancialData. Sains
Malaysiana, 44(7), 1033–1039.
[7] Joseph,R.R.,(1981)MIXTURESOFNORMALDISTRIBUTIONSANDTHE
IMPLICATIONSFOROPTIONPRICING.
[8] Kon,S.J.,(1984)ModelsofStockReturns:Finance,39,147-165.
[9] Kon,S.,(1984)ModelsofStockReturns-AComparison. The JournalofFinance,
39, 147-165.
[10] Macbeth,J.D.,Merville,L.J.,(1979)AnempiricialexaminationofBlack-Scholes
call optionpricingmodel. The JournalofFinance, 34,1173-1186.
[11] Näsholm,A.,Bunjaku,B.,(2010)ForecastingVolatility-AComparisonStudyof
ModelBasedForecastsandImpliedVolatility.
[12] Neumann,N.,(1997)OptionPricingundertheMixtureofDistributionsHypothesis.
DiskussionspapierNr.208.
[13] Ornthanalai,C.,(2014)Lévyjumprisk:Evidencefromoptionsandreturns. Journal
of FinancialEconomics, 112, 69-90.
[14] Polson,N.,Johannes,M.,(2006)MCMCMethodsforContinuous-TimeFinancial
Econometrics.
[15] Poon,S.H.,(2005)APracticalGuidetoForecastingFinancialMarketVolatility.
[16] Popovi´c,B.V.,Cordeiro,G.,Ortega,E.M.,Pascoa,M.A.R.,(2017)Anewex-
tended mixturenormaldistribution. MATHEMATICALCOMMUNICATIONS, 22,
53–73.
[17] Redner,R.A.,Walker,H.F.,(2009)MixtureDensities,MaximumLikelihoodand
the EmAlgorithm.In SIAM Review, Vol.26,195-239.
[18] Tomczak,J.M.(2012)FisherinformationmatrixforGaussianandcategoricaldis-
tributions.
[19] Xu,D.,Wirjanto,T.S.,(2009)TheApplicationsofMixturesofNormalDistributions
in EmpiricalFinance:ASelectedSurvey.
[20] Zucchini,W.,MacDonald,I.(2009)HiddenMarkovModelsforTimeSeries.