有各式各樣的參數與非參數期貨定價模型被廣泛應用於金融領域。其中一些模型的組合能顯著提升期貨定價的準確性。更具體的說，可以先通過參數模型擬合數據，再使用非參數模型學習並修正誤差估價誤差。本論文作為范和Mancini(2009) 結果的延伸，將市場交易的歐式期權價格作為輸入數據，運用「有參數模型指導的非參數定價方法」對障礙期權進行估價。「自動誤差修正估價法」運用非參數方法對由參數估價法產生的誤差進行修正，使得障礙期權的非參數定價模型可以被視為一系列的歐式期權定價的組合。在整個障礙期權的估價過程中，本論文同時提供了一種分數階快速傅裡葉變換的應用，可通過由非參數方法獲得的標的資產對數的存活函數計算標的資產對數最大值分佈的特徵函數。 / There are a variety of parametric and nonparametric option pricing models commonly used in Finance. A combination of them can enhance the pricing performance significantly. Specifically, one proposes to fit the data with a parametric method and then correct the pricing errors empirically with a nonparametric learning approach. This thesis extends Fan and Mancini's (2009) model-guided nonparametric method to barrier option pricing using market traded European option data. Adopting automatic correction of errors (ACE) method to estimate the risk neutral conditional survivor function, by which the pricing error of the initial parametric estimates is captured nonparametrically, enables the nonparametric pricing procedure to value a barrier option as a sum of sequence of European options. As a byproduct from the valuation process, this thesis also provides a modified fractional fast Fourier transform technique compute the characteristic function of the running maximum log-price of the underlying asset nonparametrically through the calibrated survivor functions. / Detailed summary in vernacular field only. / Chi, Chengzhan. / Thesis (M.Phil.)--Chinese University of Hong Kong, 2013. / Includes bibliographical references (leaves 38-39). / Abstracts also in Chinese. / Chapter 1 --- Introduction --- p.1 / Chapter 2 --- Nonparametric Local Regression Modelling --- p.4 / Chapter 2.1 --- Function Estimation by Local Constant --- p.4 / Chapter 2.2 --- Function Estimation by Local Linear Regression --- p.5 / Chapter 3 --- Nonparametric ACE European Option Pricing --- p.7 / Chapter 3.1 --- European Option Prices and Risk Neutral Survivor Functions --- p.7 / Chapter 3.2 --- Estimation of Risk Neutral Survivor Functions --- p.10 / Chapter 3.2.1 --- Risk Neutral Survivor Functions and Traded Options --- p.10 / Chapter 3.2.2 --- Survivor Function Estimation with Nonparametric ACE Method --- p.11 / Chapter 3.3 --- Representation of European Option Prices at Log-asset Level and Numerical Example --- p.15 / Chapter 4 --- Nonparametric ACE Barrier Option Pricing Framework --- p.20 / Chapter 4.1 --- Continuous-time Barrier Option --- p.20 / Chapter 4.2 --- Discrete Approximation and Backward Induction --- p.21 / Chapter 4.3 --- Decomposed Problems --- p.25 / Chapter 5 --- Nonparametric Estimation of Cumulative Distribution Function of M{U+2C7C}(R{U+209C}) --- p.28 / Chapter 5.1 --- Survivor Functions and Maxima Probabilities --- p.28 / Chapter 5.2 --- Characteristic Functions of Maxima --- p.30 / Chapter 5.2.1 --- Algorithm --- p.30 / Chapter 5.2.2 --- Preparation --- p.31 / Chapter 5.2.3 --- Fast Fourier Transform (FFT) --- p.31 / Chapter 5.2.4 --- Fractional Fast Fourier Transform (FRFT) --- p.33 / Chapter 5.2.5 --- Derivation of ΦR{U+209C} --- p.34 / Chapter 5.3 --- Numerical Experiments --- p.35 / Chapter 6 --- Conclusion --- p.37 / Bibliography --- p.38
| Identifer | oai:union.ndltd.org:cuhk.edu.hk/oai:cuhk-dr:cuhk_328731 |
| Date | January 2013 |
| Contributors | Chi, Chengzhan., Chinese University of Hong Kong Graduate School. Division of Risk Management Science. |
| Source Sets | The Chinese University of Hong Kong |
| Language | English, Chinese |
| Detected Language | English |
| Type | Text, bibliography |
| Format | electronic resource, electronic resource, remote, 1 online resource (v, 39 leaves) : ill. (some col.) |
| Rights | Use of this resource is governed by the terms and conditions of the Creative Commons “Attribution-NonCommercial-NoDerivatives 4.0 International” License (http://creativecommons.org/licenses/by-nc-nd/4.0/) |
Page generated in 0.0019 seconds