隨機波動模型下參數之最大概似估計; Maximum Likelihood Estimation of Parameters on the Stochastic Volatility Model

Please use this identifier to cite or link to this item: http://ir.lib.ncu.edu.tw/handle/987654321/43884

Title:	隨機波動模型下參數之最大概似估計;Maximum Likelihood Estimation of Parameters on the Stochastic Volatility Model
Authors:	張家榮;Chia-jung Chang
Contributors:	統計研究所
Keywords:	有母數拔靴法;蒙地卡羅模擬;Heston 模型;波動度;Heston model;volatility;Monte Carlo simulation
Date:	2010-07-02
Issue Date:	2010-12-08 14:24:10 (UTC+8)
Publisher:	國立中央大學
Abstract:	假如金融市場上的股價（stock）走勢跟隨Heston 模型，則可以藉由股價資料來推估參數，而若引進選擇權（option）的資料，來處理觀察不到的波動度，也可以由股價跟選擇權的資料推估參數。在這篇論文裡我們有兩種方法去推估參數：方法一是只有股價資料時，這時候假設波動度（volatility）是可以觀察到的；方法二是當波動度是觀察不到時，我們選用選擇權資料利用轉換來代替觀察不到的波動度去推估參數。因此我們想知道上述兩種參數估計的差距。本文使用最大概似估計法（maximum likelihood estimator, MLE）來驗證。也利用蒙地卡羅模擬以及有母數拔靴 (parametric bootstrap) 重複抽樣方法去比較兩者的結果是否一致。 In this paper, we want to know if the financial market stock price trend to follow Heston model, we can estimate the parameters by stock price data when the volatility is assumed to be observable, and while the volatility does not observed, we can use option data to deal with that, then use stock and option data to estimate the parameters. In this paper, we have two methods to estimate the parameters: one method only stock information, the assumption that volatility is to be observed; method 2 is the assumption that volatility is not observable, we use options data instead of using conversion can not see the volatility to estimate the parameters. We would like to know the difference between the estimation accuracy of the theoretical exact value. In order to count the value of this theory, we use the maximum likelihood estimator(MLE) to estimate the parameters of the stochastic volatility model in the consistency of MLE estimation. We also use the Monte Carlo simulation and the papametric bootstrap method of repeated sampling to compare the results.
Appears in Collections:	[Graduate Institute of Statistics] Electronic Thesis & Dissertation

Files in This Item:

File	Description	Size	Format
index.html		0Kb	HTML	992	View/Open

社群 sharing

Loading...